Chemical and biological factors influencing heavy metal mobilisation in the rhizosphere: implications for phytoremediation Dissertation – Doctoraatsproefschrift zur Erlangung des akademischen Grades doctor rerum naturalium (Dr. rer. nat.) ~ tot het behalen van de graad van Doctor in de Wetenschappen, richting biologie vorgelegt dem Rat der Chemisch-Geologischen Fakultät der Friedrich-Schiller-Universität Jena Voorgelegt aan de Faculteit der Wetenschappen van de Universiteit Hasselt Von / door Dipl.-Ing. Tsilla Boisselet geboren am 24. August 1983 in Périgueux, Frankreich Geboren op 24 Augustus 1983 in Périgueux, Frankrijk ISBN: 9789089130204
173
Embed
Chemical and biological factors influencing heavy metal ... · 1 Chemical and biological factors influencing heavy metal mobilisation in the rhizosphere: implications for phytoremediation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Chemical and biological factors
influencing heavy metal mobilisation in
the rhizosphere:
implications for phytoremediation
Dissertation – Doctoraatsproefschrift
zur Erlangung des akademischen Grades
doctor rerum naturalium (Dr. rer. nat.)
~
tot het behalen van de graad van
Doctor in de Wetenschappen, richting biologie
vorgelegt dem Rat der Chemisch-Geologischen Fakultät
der Friedrich-Schiller-Universität Jena
Voorgelegt aan de Faculteit der Wetenschappen van de Universiteit Hasselt
Von / door
Dipl.-Ing. Tsilla Boisselet
geboren am 24. August 1983 in Périgueux, Frankreich
Geboren op 24 Augustus 1983 in Périgueux, Frankrijk
ISBN: 9789089130204
2
Examiners / Gutachter:
1. Prof. Dr. Georg Büchel
2. Prof. Dr. Jaco Vangronsveld
Date of the public defence / Tag der öffentlichen Verteidigung: 15.06.2012
3
Selbständigkeitserklärung
Ich erkläre, dass ich die vorliegende Arbeit selbständig angefertigt habe und nur unter
Verwendung der angegebenen Hilfsmittel, persönlichen Mitteilungen und Quellen angefertigt
habe.
Jena, den 08. Mai 2012
4
The following work was developed in the context of the PhD Program carried on at the
institute of Earth Sciences of the Friedrich Schiller Universtity in Jena, Germany, under the
supervision of Prof. Dr. Georg Büchel, together with the Centre of Environmental Sciences of
the University of Hasselt, Belgium, under the supervision of Prof. Dr. Jaco Vangronsveld.
Das Forschungs-Projekt in dieser Dissertation wurde am Lehrstuhl für Angewandte Geologie
der Chemisch Geowissenschaftlichen Fakultät der Friedrich-Schiller-Universität Jena unter
der Leitung von Prof. Dr. Georg Büchel angefertigt, gemeinsam mit dem Zentrum für
Umweltkunde der Biologischen Fakultät der Universität.Hasselt unter der Leitung von Prof.
Dr. Jaco Vangronsveld.
The author of this work received a PhD fellowship from the Jena School of Microbial
Communication (JSMC).
Der Autor dieser Arbeit erhielt ein Stipendium der Jena School of Microbial Communication
(JSMC).
Financial support was given by the German Research Foundation through the program of the
Excellence Initiative.
Die finanzielle Förderung erfolgte durch die Deutsche Forschungsgemeinschaft im Rahmen
der Exzellenzinitiative.
5
Aknowledgments - Danksagung
Completing this thesis would not have been possible without the contribution of many people.
First, I would like to thank Georg Büchel for accepting me as PhD student in his group and for guiding my
studies in these challenging years. I am especially thankful for giving me the freedom to pursue my research
ideas and for introducing me to the teaching activities. More importantly, I am grateful for his sincerity, his
confidence and his encouraging discussions.
To Dirk Merten, for taking the time to read all manuscripts, having an open ear, and especially for leading
me at the beginning when I still needed orientation. My appreciation words include also all my (former)
colleagues of the group of Applied Earth Sciences, Andreas, Christian, Martin, Franziska, Anika, Anja,
Delphine, Daniel, Sisay and Anahita for making my integration in Jena and in the working group easier. I
express my gratitude to my collegues for their useful comments about my presentations and on the
provisional results, and to all the co-authors of the manuscripts where I have been involved in, for sharing
their experience with me. My special thanks to Rochus Merker for his cure of my computer and my laptop!
I wish to express my great acknowledgments to Jaco Vangronsveld for accepting me at his Institute and
offering good working conditions and facilities. I would like to thank for suggesting me the cotutelle
agreement, and for his valuable comments and suggestions for my research his has given. I want to thank at
this place also Nele Weyens for her advises during the experimental trials, and the working group at the
center for environmental sciences for their help and friendliness at any time. Bedankt Michiel, Sarah, Els,
Bram, Sophie, An, Jan en Sasha!
I wish to acknowledge the members of the JSMC, especially Ulrike Schleier and Carsten Thoms, for giving
me the opportunities of the JSMC network and all seminars and courses offered, and for giving me the
chance and the flexibility to go abroad to continue my project.
I am deeply indebted to the persons working at the Thüringer Landesanstalt für Landwirtschaft, especially
Hr. Schröter and his co-workers for the practical and useful advice to conceive pot experiments and for the
cure during the growth.
I owe a special thanks to the Hydrology lab team at the Institute of Applied Earth Sciences for their advice
and support for all analytical work, Dirk Merten, Ulrike Buhler, Ines Kamp and Gerit Weinzierl.
I thank also all co-workers at the Institute for Microbiology in the group of Erika Kothe, especially Götz,
Eleen, Francesca, Martin, for sharing their experience and lab space with me.
I would also like to thank Björn Grübler from the Institute of Botanics, Plant Physiology for the sharing his
knowledge in chlorophyll fluorescence measurements, and Elke-Martina Jung for kindly sharing her
expertise for the microscopy. I would like to acknowledge the contribution of Sandra Studenik from Institute
of applied and ecological microbiology for helping and letting me performing one experiment in her lab.
I would like to acknowledge the students Sebastian Wagner, and Marcus Meyer for their contribution and
especially Claudia Rebentisch for her enthusiastic participation during the last experiment in Jena.
I would also like to thank Jan Meiling from the Wageningen University for his help in providing me ideas
for cooperation institutes, and through whom I got to know Hasselt University.
I would like to address my gratitude to the German Research Foundation (DFG) and the JSMC for
providing the financial support to accomplish this scientific investigation.
Finally, I am deeply grateful to my family for being close to me even from the distance and for following
enthusiastically the experiences I shared with them. Thank you to my parents and to Christiane and Agnès
for reviewing the translation of the abstract!
A very special thank goes to all my friends I met here in Jena, for sharing my life during this years, for your
precious presence and the joyful moments that enriched my soul and softer the difficult moments. Thank
I cannot end without mentioning Raffaele, the linking point that provoked the deep association and tight
link between private life, travelling, linguistics and scientific research from the very first day, strongly
contributing to make my past three and a half years my most challenging and exciting… till now!
Thanks God, Thank you, Merci, Danke,
Tsilla Boisselet
Jena, im Mai 2012
6
À ceux que j’aime, à ceux qui m’ont appris la valeur de la solidarité
“L’union fait La force”
Concordia parvae res crescunt,
discordia maximae dilabuntur,
Sallust (Jugurtha, 10)
7
ABSTRACT
The thesis aims at investigating the potential of phytoremediation on a heavy metal
contaminated soil with very low nutrient content, low organic carbon and acidic pH. The soil
originates from the Ronneburg mining district in Thuringia (Germany) which was the third
largest uranium-producing area worldwide. The mining activities strongly altered the
hydrogeology of the area. The acidic and highly mineralised solutions caused by leaching of
waste heaps infiltrated into the soil and underlying sediments and polluted the water-soil
system with high concentrations of Mn, Al, Ni, U, and Rare Earth Elements (REE). Despite
remediation activities since the 1990s, contamination is still measurable. Since the soil pH is
quite low (pH 4-4.5), the mobility and bioavailability of trace elements is high and so the
amounts taken up by plants are significant. So, to study the interaction between soil trace
elements and plants, in particular via root exudates, four plant species were chosen: Triticale
(× Triticosecale), sunflower (Helianthus annuus), red fescue (Festuca rubra) and red clover
(Trifolium pratense), grown as monoculture and polyculture. The last two were used for
microbial studies, including the isolation and characterisation of endophytes potentially useful
for remediation enhancement.
The substrate at the study area has been extensively characterised, and sequential extraction
already allowed predictions about possible bioavailability of metals. However, the active
influence of plants and their root exudates were not taken into consideration. Therefore, in a
first part, REE will be used as a way to study root impact on element mobilisation, by
comparing leaching by different organic and inorganic solutions. REE form a consistent
group of so called metals, whose pattern, resulting from normalisation to a standard, can be
used to describe different processes of dissolution and preferential precipitation. Our study
shows that metals (REE) are mobilised in a different way by acidic solutions of different
origin, and that organic acids lead to a different fractionation than inorganic ones. REE
pattern changes were also observed in plants and their rhizosphere. The amounts of soluble
trace elements decreased in the rhizosphere zone, while pH increased. Based on the analysis
of REE patterns, it seemed that organic substances, like organic acids were an important
factor that mobilises metals in the rhizosphere and allows their uptake into the plant.
Furthermore, combined cultivation generally had a beneficial effect on plant growth; plants
showed later necrosis and had a higher biomass production in relation to the initial seed
quantity. Plants also had a clear effect on the soil structure: especially clover and red fescue
were producing extended root networks, holding the soil. Festuca especially retained water.
These features were considered to be interesting for remediation in sites with an erosion risk.
Microorganisms living around and inside the plants also influence their growth and mineral
uptake. If more is known about these, they can be used, if chosen well, to enhance
phytoremediation processes, especially on soil poor in nutrients. In the present study we
concentrated on bacteria living inside the plant tissues, and isolated, characterised and finally
identified the cultivable ones. 78 stable, morphologically distinct isolates were obtained,
belonging to 32 genera, although 12 isolates could not be identified. The identified
endophytic community was different for the 2 studied plants, so it seems that a selection took
place. The endophytic bacteria showed additionally clear spatial compartmentalisation within
the plant, suggesting that they can form specific associations with plant tissue. Furthermore,
the specificity of some strains for some compartments suggests that different uptake
8
mechanisms for different plant tissues exist. They were found to be more diverse in the upper
parts of the plants. Nevertheless, several strains isolated from roots could not be identified.
Many of the isolated genera are very similar to known plant endophytes, and a large number
of them are also related to strains used to support phytoremediation, mostly on sites
contaminated with organic pollutants.
A number of isolates demonstrated the capacity to produce plant growth promoting
substances and resistance to the trace elements enriched in the contaminant soil. As a
consequence, some of these strains were used to promote growth of Festuca rubra and
Trifolium pratense, and were inoculated separately to each plant and also as bacterial
consortia of 2 or 3 strains. The inoculated plants showed better growth, higher plant density,
healthier appearance, better and denser developed root network which, by consequence, was
leading to a better soil structure. Moreover, the inoculated plants showed a higher
photosynthetic efficiency, which can be interpreted as an improved fitness due to a better
stress resistance. Further, the positive effect of the bacteria is enhanced in case consortia of
strains are used. The effect of bacteria on trace element mobility and metal uptake depends
mainly on the element itself: for instance Al was less present in the soluble fraction of the
soil, and Mn more mobile in the soil after the combined action of plants and microbial
consortia. Zn on the other hand was not influenced.
As the studied plants have a clear influence on metal mobility and pH, it is useful to use
wisely their properties for remediation purposes. There is a large number of symbiotic
bacteria described, which are living inside their tissue, and a notable part of them show
promising properties for the support of plant growth and remediation. We suggest therefore
using Festuca and Trifolium as complement to extracting, hyper-accumulating plants or to
stabilising plants, in order to increase soil fertility and protection of soil erosion via the dense
root network. Festuca is more influenced by bacteria concerning its root development, so
should therefore get particular attention when it comes to choosing plant populations for
remediation. It is also of importance to combine plants of different species to ensure long-
term system stability.
9
ZUSAMMENFASSUNG
Diese Arbeit befasst sich mit der Untersuchung des Potenzials für Phytoremediation auf
einem mit Schwermetallen kontaminierten Boden mit sehr geringem Nährstoffgehalt,
niedrigem organischem Kohlenstoffgehalt und saurem pH-Wert. Der Boden stammt aus dem
ehemaligen Bergbaugebiet Ronneburg in Thüringen (Deutschland), das der drittgrößte
Uranproduzent weltweit war. Die Bergbauaktivitäten veränderten tiefgehend die
Hydrogeologie des Gebietes. Die sauren und stark mineralhaltigen Abwässer, die durch
Auslaugen der Halden entstanden, sickerten in den Boden und kontaminierten das Wasser-
Boden-System mit hohen Mengen an Mangan, Aluminium, Nickel, Uran sowie Seltenen
Erdelementen (SEE). Trotz umfangreicher Sanierungsaktivitäten seit den 1990er Jahren xx ist
die Kontamination noch an vielen Stellen messbar. Da der Boden-pH ziemlich sauer ist (pH
4-4,5), ist die Mobilität und Bioverfügbarkeit von Spurenelementen hoch und die von
Pflanzen aufgenommen Mengen signifikant. Um die Wechselwirkung zwischen
Bodenelemente und Pflanzen zu untersuchen, insbesondere durch die Wurzelexudate, wurden
vier Pflanzenarten ausgewählt: Triticale (× Triticosecale), Sonnenblumen (Helianthus
annuus), Rotschwingel (Festuca rubra) und Rotklee (Trifolium pratense), die als Monokutur
sowie als Polykultur kultiviert wurden. Nur die beiden letzten Pflanzenarten wurden für
spätere mikrobiologische Untersuchungen verwendet, insbesondere für die Isolierung und
Charakterisierung von potentiell nützlichen Endophyten in Hinblick auf Sanierungszwecke.
Das Substrat wurde umfassend charakterisiert und mithilfe sequentieller Extraktion konnten
bereits Aussagen über den bioverfügbaren Metallanteil getroffen werden. Allerdings wurde
der aktive Einfluss von Pflanzen und deren Wurzelausscheidungen nicht berücksichtigt.
Daher werden im ersten Teil SEE zur Hilfe gezogen, um den Einfluss von Wurzelexudaten zu
untersuchen, indem Elutionen mittels verschiedener organischer sowie anorganischer
Lösungen und die daraus entstandenen SEE Muster verglichen werden. SEE bilden eine
konsistente Gruppe von so genannten Metallen, deren Muster, das sich aus der
Normalisierung zu einem Standard ergeben, für die Beschreibung unterschiedlicher Lösungs-
und Präzipitationsprozesse benutzt werden kann. Unsere Studie zeigt, dass Metalle (inkl.
SEE) in unterschiedlicher Weise durch verschiedene saure Lösungen mobilisiert werden und
dass organischen Säuren zu einer anderen Fraktionierung führen als anorganische. SEE-
Muster Veränderungen wurden auch in den Pflanzen und in ihrer Rhizosphäre beobachtet.
Die Menge löslicher Spurenelemente nahm in der Rhizosphärenzone ab, während der pH-
Wert zunahm. Basierend auf der Analyse von SEE-Mustern scheint es, dass organische
Substanzen wie organische Säuren ein wichtiger Faktor sind, der Metalle in der Rhizosphäre
mobilisiert und deren Aufnahme in die Pflanze ermöglicht.Weiters ergab sich, dass die
Polykultur einen positiven Effekt auf die Pflanzen hatte; sie zeigten später Nekrosen und
hatten eine höhere Biomasseproduktion in Bezug auf die ursprüngliche Menge Samen. Die
Pflanzen hatten auch eine deutliche Wirkung auf die Bodenstruktur: vor allem Klee und
Rotschwingel produzierten ein stark ausgebildetes Wurzelnetzwerk, das den Boden festigt;
insbesondere Festuca konnte dadurch viel Wasser zurückhalten. Diese Eigenschaften sind vor
allem für die Sanierung von Standorten mit Erosionsgefährdung wichtig.
Die Pflanzen und ihre Spurenelement-Aufnahme können durch viele andere Faktoren
beeinflusst werden, da sie einen eigenen Mikrokosmos in ihrer Rhizosphäre bilden.
10
Mikroorganismen, die um und in der Pflanze leben, beeinflussen ebenfalls deren Wachstum
und Mineralstoff-Aufnahme. Wenn mehr über diese bekannt ist, können sorgfältig
ausgewählte unter ihnen verwendet werden, um Phytosanierungsprozesse, inbesondere auf
nährstoffarmen Böden zu verbessern. In der vorliegenden Studie wurde der Fokus auf
Bakterien, die innerhalb des Pflanzengewebes leben, gerichtet und die kultivierbaren unter
ihnen wurden isoliert, charakterisiert und schließlich identifiziert. Es ergaben sich 78 stabile,
morphologisch unterschiedliche Isolate, aus 32 Gattungen; 12 Isolate konnten aber nicht
identifiziert werden. Die identifizierte endophytische Population war unterschiedlich für die 2
untersuchten Pflanzen, anscheinend fand eine Selektion statt. Die endophytischen Bakterien
zeigten außerdem eine klare räumliche Trennung innerhalb der Pflanze, was darauf hindeutet,
dass sie charakteristische Assoziationen mit bestimmten pflanzlichen Geweben bildeten.
Darüber hinaus deutete die Spezifität einiger Stämme für bestimmte Kompartimente auf
unterschiedliche Aufnahmemechanismen für unterschiedliche Pflanzengewebe hin. Die
Diversität war größer in den oberen Pflanzenteilen. Allerdings konnten mehrere Stämme aus
den Wurzeln nicht identifiziert werden. Viele der gefundenen Gattungen sind bekannten
Pflanzenendophyten ähnlich und viele von ihnen werden auch verwendet, um Phytosanierung
zu unterstützen, vor allem auf Standorten, die mit organischen Kontaminanten belastet sind.
Eine beachliche Anzahl der Isolaten zeigte Resistenz gegen toxische Metalle, die in dem
Substrat vorhanden sind, sowie die Fähigkeit, Pflanzenwachstum fördernde Substanzen zu
bilden. Daher wurden einige dieser Stämme verwendet, um das Wachstum von Festuca
rubra und Trifolium pratense zu fördern; sie wurden einzeln sowie als Konsortien von 2 oder
3 Stämmen inokuliert. Die inokulierten Pflanzen zeigten ein besseres Wachstum, eine höhere
Pflanzendichte, gesünderes Aussehen, ein besser und dichter entwickeltes Wurzelsystem, das
zu einer besseren Bodenstruktur führte. Die inokulierten Pflanzen zeigten außerdem eine
höhere photosynthetische Effizienz, die als eine verbesserte Stressresistenz interpretiert
werden kann. Weiters ist die positive Wirkung der Bakterien erhöht, wenn mikrobielle
Konsortien verwendet werden. Die Wirkung von Bakterien auf Spurenelement-Mobilität und
Metallaufnahme hängt vor allem von dem Element selbst ab: zum Beispiel war Aluminium
nach der kombinierten Wirkung von Pflanzen und mikrobiellen Konsortien in geringeren
Mengen in der löslichen Fraktion des Bodens vorhanden und Mangan im Gegenteil mobiler
im Boden. Zink andererseits wurde nicht beeinflusst.
Da die verwendeten Pflanzen eine klare Wirkung auf die Metallmobilität und den pH-Wert
zeigen, ist es von Vorteil diese Eigenschaften gezielt zu nutzen. Eine große Anzahl an
endosymbiotischen Bakterien wurde beschrieben und ein großer Anteil davon zeigt
vielversprechende Eigenschaften für die Verbesserung von Pflanzenwachstum und
Phytoremediation. Wir empfehlen daher die Verwendung von Festuca und Trifolium als
Ergänzung zu extrahierenden, Metall-Hyperakkumulator Pflanzen, oder zu stabilisierenden
Pflanzen, um einerseits die Bodenfruchtbarkeit zu erhöhen und andererseits als
Erosionsschutz wegen des dichten Wurzelwerks. Am meisten wird die Wurzelentwicklung
von Festuca durch Bakterien beeinflusst, daher sollte bei der Auswahl von Pflanzenarten für
die Sanierung dieser Pflanze besondere Aufmerksamkeit gewidmet werden. Es könnte auch
hilfreich sein, Pflanzen verschiedener Arten zu kombinieren, um eine langfristige Stabilität
des Systems zu gewährleisten.
11
RÉSUMÉ
La thèse vise à étudier le potentiel de la dépollution par les plantes (phyto-assainissement) sur
un sol contaminé par des métaux lourds, pauvre en éléments nutritifs, avec une faible teneur
en carbone organique et un pH acide. Le sol provient de l’ancienne exploitation minière du
district de Ronneburg, en Thuringe (Allemagne), qui était le troisième producteur mondial
d'uranium. Les activités minières ont profondément altéré l'hydrogéologie de la région. Les
solutions acides et fortement minéralisées, produites par la lixiviation des terrils, se sont
infiltrées dans le sol, et ont pollué le système eau-sol avec des quantités élevées d'uranium,
Terres Rares et autres éléments toxiques, principalement Mn, Al et Ni. Malgré les activités
d'assainissement entreprises depuis les années 1990, le niveau de contamination est toujours
mesurable dans plusieurs endroits du site. Le pH du sol étant assez bas (pH 4-4,5), la mobilité
et la biodisponibilité des métaux sont élevées et les quantités absorbées par les plantes sont
importantes. Afin d’étudier l'interaction entre les oligo-éléments du sol et les plantes, en
particulier à travers les exsudats racinaires, quatre espèces de plantes ont été choisies: le
triticale (× Triticosecale), le tournesol (Helianthus annuus), la fétuque rouge (Festuca rubra)
et le trèfle violet (Trifolium pratense), cultivées en monoculture et polyculture. Seules les
deux dernières ont été utilisées pour les examens microbiologiques, incluant l’isolation et la
caractérisation de bactéries endophytes potentiellement utiles pour l’amélioration de
l’assainissement.
Le sol de la zone d'étude a été largement caractérisé, et l'extraction séquentielle a été utilisée
pour estimer la biodisponibilité potentielle de certains éléments toxiques. Cependant,
l'influence active des plantes et leurs exsudats racinaires n’ont pas été pris en considération.
Par conséquent, dans une première partie, les Terres Rares sont utilisés comme outil pour
étudier l’impact des racines sur la mobilisation des métaux, en comparant la lixiviation du sol
avec différentes solutions organiques et anorganiques. Les Terres Rares forment un groupe
cohérent de métaux, et sont généralement graphiquement représentés sur un diagramme de
distribution obtenu après la normalisation à un standard. Le motif varie en fonction des
conditions physico-chimiques du système, et peut être utilisé pour décrire différents procédés
notamment de dissolution et de précipitation.
Notre étude montre que les métaux (dont les terres rares) sont mobilisés d'une manière
distincte par différentes solutions acides, et que les acides organiques conduisent à un
fractionnement spécifique de celui causé par les acides inorganiques. Les terres rares ont
également montré des changements de leur motif dans les plantes et leur rhizosphère. La part
de métaux solubles a été diminuée dans la zone rhizosphère, tandis que le pH a été augmenté.
En se basant sur l'analyse des motifs de terres rares, il semble que les substances organiques
comme par exemple des acides organiques ont été un facteur important pour la mobilisation
des métaux dans la rhizosphère et par conséquent pour leur absorption dans la plante. De plus,
la polyculture s’est montrée bénéfique pour les plantes: elles montrent une nécrose plus
tardive, et ont une production supérieure de biomasse par rapport à la quantité de semis
initiale. Cela est particulièrement visible pour le trèfle. Les plantes ont aussi eu un effet
manifeste sur la structure du sol: en particulier le trèfle et la fétuque rouge ont produit un
réseau de racines longues qui tient le sol. Particulièrement avec Festuca la retenue d'eau était
importante. Ces caractéristiques sont intéressantes pour l'assainissement dans les sites
présentant un risque d'érosion.
12
De nombreux organismes vivant autour des plantes peuvent influencer leur croissance et
l’assimilation de minéraux. Une connaissance plus approfondie de ces micro-organismes et de
leurs propriétés permettrait de les utiliser, choisis judicieusement, pour améliorer les
techniques d’assainissement par les plantes, en particulier sur des sols pauvres en nutriments.
Nous nous sommes concentrés dans la présente étude sur les bactéries vivant à l'intérieur des
tissus végétaux (endophytes), et avons isolé, caractérisé et enfin identifié les endophytes
cultivables. 78 isolats stables, morphologiquement distincts ont été obtenus, appartenant à 32
genres; 12 isolats n'ont pas pu être identifiés. La communauté endophyte identifiée était
différente pour les 2 plantes étudiées, il semble donc qu’une sélection spécifique pour chaque
espèce ait eu lieu. Les bactéries endophytes ont montré en outre, clairement la
compartimentation spatiale au sein de la plante, ce qui suggère qu'elles peuvent former des
associations caractéristiques avec certains tissus végétaux. Ensuite, la spécificité de certaines
souches pour certains compartiments suggère qu’il existe des mécanismes d’assimilation
divergents pour différents tissus végétaux. La diversité était plus élevée dans les parties
supérieures des plantes. Néanmoins, plusieurs souches isolées des racines n'ont pas pu être
identifiées. La plupart des genres sont connus pour être des endophytes de plantes, et
beaucoup d'entre eux sont également utilisés pour améliorer le phyto-assainissement, surtout
sur les sites contaminés par des polluants organiques.
Un certain nombre d'isolats ont démontré la capacité de produire des substances favorisant la
croissance végétale et la résistance aux métaux toxiques présents dans le sol. En conséquence,
certaines de ces souches ont été utilisées pour promouvoir la croissance de Festuca rubra et
Trifolium pratense, et ont été inoculées séparément pour chaque plante ainsi que par
consortiums bactériens de 2 ou 3 souches. Les plants inoculés ont montré une meilleure
croissance, une densité de plants par pot supérieure, une apparence plus saine, un réseau de
racines mieux développé et plus dense, qui a, par conséquent, conduit à une meilleure
structure du sol. Les plantes inoculées ont montré une plus grande efficacité photosynthétique,
qui peut être interprétée comme une meilleure santé en raison d'une résistance supérieure au
stress. En outre, l'action positive des bactéries est renforcée dans le cas d’inoculation de
consortiums de souches. L'effet des bactéries sur la mobilité l'absorption des métaux dépend
surtout de l'élément lui-même: par exemple, l’aluminium était moins présent dans la fraction
soluble du sol, et le manganèse plus mobile dans le sol après l'action combinée des plantes et
des consortiums microbiens. La solubilité du zinc, d'autre part n'a pas changé.
Comme les plantes étudiées ont une influence évidente sur la mobilité des métaux et le pH, il
est utile d'utiliser judicieusement leurs propriétés pour la dépollution. Un grand nombre de
bactéries symbiotiques vivant à l'intérieur de leurs tissus ont été décrites, et une partie
importante d'entre elles présente des propriétés prometteuses pour le soutien de la croissance
des plantes et l'assainissement. Nous suggérons donc d'employer Festuca et Trifolium en tant
que complément à des plantes utilisées pour l'extraction, notamment des plantes hyper-
accumulatrices, ou bien des plantes stabilisatrices, afin d'accroître la fertilité du sol et comme
protection contre l'érosion, en raison de leur réseau racinaire dense. Festuca est plus
influencée par des bactéries concernant le développement de ses racines, et devrait donc faire
l’objet d’une attention particulière dans l’avenir quand il s'agira de choisir les populations de
plantes pour l'assainissement. Il est également important de combiner des plantes de
différentes espèces, pour assurer la stabilité du système à long terme.
13
TABLE OF CONTENTS
GENERAL INTRODUCTION, PRESENTATION OF THE WORK
1 Background and motivation .......................................................................................... 18
1.1 History of the study area ......................................................................................... 18 1.2 Characterisation of the studied substrate ................................................................. 20
1.3 Phytoremediation- principles and challenges .......................................................... 23
2 Root exudates: the active contribution of plants to metal mobility ................................. 26
3 Rare earth elements – a tool for tracing ......................................................................... 30
4.1 Description of the endophytic population in two plants grown on this specific
contaminated substrate ...................................................................................................... 34 4.1.1 Identification of the strains .............................................................................. 35
4.1.2 Spatial distribution of the strains and bacterial community .............................. 35 4.2 Improving plant growth on heavy metal contaminated soil using selected endophytic
microorganisms ................................................................................................................ 36 4.2.1 Plant microorganism partnerships for a better biomass production ................... 36
4.2.2 Selection of strains for phytoremediation support ............................................ 38
1.3 Sorption on minerals ....................................................................................... 53
2 Root exudates: the active contribution of plants to metal mobility ................................. 58
2.1 Local dependence of the acid concentration ............................................................ 60 2.2 Impact of nutrients.................................................................................................. 62
2.3 Complexation by organic acids ............................................................................... 63 2.4 Microbial influence – symbiotic microorganisms .................................................... 63 2.5 Other factors ........................................................................................................... 64
3 Microbial influence on metal mobility .......................................................................... 66
3.1 Bioleaching – dissolving metals from rocks and ore ............................................... 67
3.1.1 Description of the process ............................................................................... 68 3.2 Precipitation of metals by microorganisms ............................................................. 69
3.3 Dissolving of nutrients and interaction with the flora .............................................. 69
4 Bioremediation: application of natural processes ........................................................... 70
4.1 Bioremediation strategies involving microorganisms or plants ................................ 71
4.2 Plant-microorganism partnerships for an improved bioremediation strategy and a
modified metal uptake ...................................................................................................... 72
14
CHAPTER 2: STUDY OF THE INFLUENCE OF PLANT ROOT EXUDATES ON HEAVY
METAL MOBILITY BY MEANS OF ANALYSIS OF RARE EARTH ELEMENT
4.1 Plant growth and biomass production ..................................................................... 97
4.2 Plant influence on element mobility in soil and soil water ....................................... 98 4.3 REE fractionation and specificity of the HREE enrichment .................................... 99
4.4 Metal uptake and REE patterns in plants ............................................................... 100 4.5 Effect of plant consortium .................................................................................... 101
2 Materials and methods ................................................................................................ 109
2.1 Collection and handling of samples ...................................................................... 109 2.2 Purification ........................................................................................................... 109
2.3 DNA extraction and PCR amplification ................................................................ 110 2.4 DNA sequencing .................................................................................................. 110
3.1 Plant growth promoting effects: macroscopic observation of roots and growth
density ............................................................................................................................ 151
3.2 Recovery of inoculated strains in plants ................................................................ 153 3.3 Chlorophyll fluorescence: plant stress................................................................... 155
3.4 Analysis of soils: water and ammonium nitrate extractable fractions of metals ..... 155 3.5 Analysis of plants: total metal content .................................................................. 157
4.1 Root growth .................................................................................................. 157 4.2 Bacterial colonisation of Festuca and Trifolium ............................................. 158
4.3 Protection against stress (Photosynthesis) ...................................................... 158 4.4 Inoculation with consortia has more positive effects than single strain inocula
159 4.5 Mobilisation of metals in the soil ................................................................... 160
4.6 Metal uptake.................................................................................................. 160
regulators and secondary metabolites. The complex interactions in the soil are the result of the
chemical interaction between the different organic compounds excreted by plant roots and the
different microscopic actors of the soil, each interacting specifically with different
compounds. Hence, each plant species have a different rhizosphere micro-flora in terms of
abundance and physiological characteristics, which can be further modified by the properties
of the soil, plant age and plant nutritional status (Marschner, 2012). [Cf. chapter 1 §2 for more
details]
Root exudates play a role in the weathering of soil, for the mobilisation of nutrients as P,
NH4+ or Fe, especially the organic acids, phytosiderophores, and phenolic compounds.
Moreover they are important for the protection of plants against uptake of heavy metals into
the roots; thereby the main agents are citrate, malate, or small peptides. Additionally phenolic
compounds, organic acids, sugars play a role for attraction of useful microorganisms. Root
exudates can also act as signal molecules or as precursors for hormones. One of the important
aspects that should be stressed in this study is their influence on trace metals. Indeed,
exudates are known to enhance with great efficiency their amounts in the bioavailable phase
of the soil and therefore in the plant, but also to modify their speciation to avoid toxic effects
by an excess of them. Very low concentrations are sufficient for their biological effect
(Marschner, 2005). The sort, composition, amounts, proportion are influenced by many
factors, as plant species, age, soil composition and so on. Since the secretion is motivated by
physiological needs of
the plants, nutrients
present in the soil have a
major impact on
exudation, usually
enhancing the process,
particularly with regard
to the supply of N, P and
K. Unlike their secretion
to attract or sustain microorganisms, the secretion for the uptake of nutrients is irregular and
rather occurring as pulses of substances release in high locally concentrations within a short
period of time (Marschner, 2012). Further, the distance to the root plays an important role.
The high organic acid concentrations can be found in the very close rhizosphere zone, and
29
Stress by lack of nutriments or metal toxicity leads to a changes behaviour of the plant
concerning the secretion of substances through the roots.
almost not at all in the bulk soil, giving to the rhizosphere very different properties. On the
other hand, there can be a parallel mobilisation and immobilisation of metals by the same
procedure, depending on the conditions. Organic acids are known to complex metals in the
same way as EDTA (Díaz-Barrientos et al., 1999), these complexes are very stable and can
enhance the availability of metals. They have also a buffer effect, that increases with the
quantity of acid (Yuan et al., 2007). This leads to a complex interaction between heavy metal
mixture in the soil, plant exudates, soil minerals and organic content in the soil.
The spectrum is quite broad, but the most common acids are malonate, citrate, malate, oxalate
and fumarate. The first four were chosen and applied as a mixture in leaching experiments to
represent the leaching occurring in the root zone. Citrate, malate and oxalate are very efficient
to dissolve metals, because they can form stable 5 or 6 membered ring structures with
trivalent ions like FeIII
and Al. That is why the most common response to Al stress,
recognizable at the characteristic inhibited root growth, is complexation through organic acids
(Barceló and Poschenrieder, 2002), most commonly citrate, malate (López-Bucio et al.,
2000), or oxalate (Barceló and Poschenrieder, 2002). The efficiency for the detoxification of
plants decreases from citrate over oxalate to malate. These phenomena are also observed for
Zn-resistant plants (Barceló and Poschenrieder, 2002). An excess of Al can inhibit the uptake
of other elements, like Ca, Mg, Zn and Mn. Some acids have a higher affinity to some
specific metals (Yuan et al., 2007): citric acid for instance is better for Cu mobilisation, oxalic
acid better for Cd mobilisation.
The capacity of dissolution of inorganic P (Pi) is thus highly correlated with the number of
OH- and COOH- functional groups and their position in the chain (high affinity to divalent
and trivalent acids). That is the reason why citrate has the highest P dissolving capacity
among common organic acids. This way of phosphate acquisition is important for plants
adapted to acid mineral soils with very low Pi availability (Grayston et al., 1996; López-Bucio
et al., 2000; Oburger et al., 2009; Shen et al., 2002). In alkaline soil and low availability of P
and Fe, many dicotyledonous plants react to the iron stress by secretion of H+ by the roots,
reduction of FeIII
to FeII, production of root exudates, mainly malate, citrate. In the meantime,
a decrease of the pH was observed.
Low concentrations, locally delimited make it difficult to detect them, especially if they are
quickly degraded by the microorganisms interacting with the plants. The quantification of
organic acids and in general of root exudation under natural conditions is difficult due to
binding of exudates to soil components, assimilation and the degradation by microorganisms
(turnover rate) under non-sterile conditions, and the lower production under sterile conditions.
The presence of microorganisms is one factor which can lead to modifications in the quality
and quantity of root exudates (Grayston et al., 1996). The stimulation of exudation occurs in
both herbaceous plants and trees. Organic acids can also be produced by microbial activity,
stimulated by the production of organic carbon and CO2 from the roots, making it difficult to
discriminate between the action of the flora and the microorganisms.
30
REE form a consistent group of
metals, whose pattern, resulting
from normalisation to a standard,
can be used to describe different
processes of dissolution and
preferential precipitation.
One method used to investigate the process of metal mobilisation in situ and to overcome the
analysis difficulties was to follow the signature given by the pattern of Rare Earth Elements
(REE).
3 RARE EARTH ELEMENTS – A TOOL FOR TRACING
3.1 Why REE? Rare Earth Elements (REE) are elements of the lanthanide (La-Lu) group, often Yttrium and
Scandium are also counted as REE. They occur as pure or mixed oxides, that are not found is
great amounts, so their name. Though, more recent analysis showed that for instance Cerium
is four times more common than Lead in the earth crust (Ferreira da Silva et al., 2009). These
metals are very similar to each other’s because of their similarity in the electron structure.
They show smooth, but continuous variations in chemical behaviour as a function of their
atomic number. They are strongly electropositive, and occur in oxidation number 3, either as
stable oxides, carbides or borides (Spedding, 2009),
therefore they are used as chemical analogues to the
trivalent actinides (Am, Cm, Cf) which are difficult to
study because of their toxicity, radioactivity and variety
of oxidation states (Ding et al., 2006). Only Ce und Eu
can be found in other valences (Ce4+
or Eu2+
) giving
them other properties depending on the redox potential.
The ion radius decreases from La3+
to Lu3+
, this
phenomenon is called lanthanides contraction. Rare Earth Elements are often separated in
light (LREE; La-Sm), middle (MREE; Sm−Dy) and heavy REE (HREE; Ho-Lu).
The pattern obtained through normalisation with a standard (here PAAS, Post Archean
Australian Shale, McLennan, 1989) is a tool to study water-rock-interactions, as tracer to find
out the erosion and formation of sediments, or to follow the flow of water and in case of
AMD influenced areas to follow contamination by heavy metals or radionuclides (Merten et
al., 2005). To follow erosion the initial rock is often taken as the standard, for instance Basalt,
or Granite (Aubert et al., 2001; Steinmann and Stille, 2008).
Different factors can influence the pattern of REE. Not only the source material causes a
typical pattern, but also pH (precipitation), redox-conditions and ligands can alter this
(Aström, 2001; Semhi et al., 2009; Shan et al., 2002). Cao et al. (2001) shows further how
lower pH and lower Redox potential lead to release of La, Ce, Gd and Y by changing their
speciation from Fe-Mn oxides In particular organic substances excreted by plants to optimise
their nutrient input can play a role in the regulation of the soil pH and as well complex
different metals (van Hees et al., 2003). The quantity of REE in solution therefore does not
depend only on the pH, but also on the presence and amount of Al, Fe and Mn. Generally, the
adsorption of REE is connected to the cation exchange capacity of the soil.
Anomalies are another aspect of the pattern. They appear mainly because of changes in the
redox conditions, which change the oxidation state of some REE, like Ce or Eu. These behave
then differently, being more easily precipitated, or at the contrary being dissolved more easily,
or showing a higher affinity for adsorption on some phases like hydroxides. Hence, Eu occurs
as Eu2+
under reducing conditions, and can be incorporated in minerals instead of Ca2+
or
31
Sr2+
. Positive Ce anomaly is said to be typical for REE fixed in oxy-hydroxides. The Ce(III)
oxidising capacity of Fe oxyhydroxide-precipitating systems is considerably higher than that
of systems in which dissolved REE interact with preformed Fe oxy-hydroxides (Bau, 1999).
Organic content plays an important role also for the fractionation of Ce. Indeed, alkaline
water with positive Ce anomaly is often poor in organics, whereas organic-rich waters show a
negative Ce anomaly, showing that Ce is bound by humic acids presents in the water (Pourret
et al., 2008). The critical pH for Ce anomaly is pH 5 (Lei et al., 2008). Ce is oxidised to CeIV
and sediments as CeO2. This leads to negative Ce anomaly in water. Gd anomalies are most
often of anthropological origin due to the use of this element in medical imaging techniques
(Möller et al., 2002; Rabiet et al., 2009).
Rare Earth Elements (REE) were chosen as a tool for several reasons. Their occurrence in all
parts of the studied system in detectable concentration was one significant advantage and
basic condition for their use as a tracer. Additionally, the availability of several data from
other studies allowed making comparisons and drawing conclusions about the causes of some
pattern characteristics. Finally, the patterns formed as a result of the different environmental
parameters can be found even after the direct cause is not detectable anymore. So, some
enrichment of certain REE in the soluble phase can be found, even if the solid precipitates
retaining the other REE are not found at the place of study, or if the substances specifically
dissolving and therefore provoking the enrichment of some REE are already degraded. Hence,
it is a way to overcome analytical difficulties.
3.2 Heavy metal uptake and REE fractionation by plants on this specific
contaminated soil
Because of the relatively high content of metal including REE and quite low pH, the mobility
and bioavailability of metals is high and so the up taken amounts of metals by plants are
significant. Therefore, this substrate is adequate for monitoring soil parameter and metal
behaviour [See chapter 2].
3.2.1 Metal leaching and REE fractionation
Therefore, in a first part, with a focus on REE, the effect of different leaching solutions was
tested; in order to describe and differentiate different metal mobilising effects of the studied
system, in particular plant exudates. REE fractionation by the studied plants
In order to come closer to the natural conditions and understand on-going processes in the
study area, two autochthonous plants (Dietrich and Berger) were chosen (Figure 6). Festuca
rubra is a very resistant grass found in many heavy metal contaminated areas, and known as a
pioneer plant. Clover (Trifolium pratense) was chosen because of its ability to fix air nitrogen
due to its symbiosis with bacteria, and so to overcome partially the nitrogen-poorness of the
soil. Additionally, reference plants, sunflower (Helianthus annuus) and Triticale (hybrid of
wheat (Triticum) and rye (Secale)), enable comparison with previous studies, especially
regarding REE fractionation (Kidd et al., 2009; Lonschinski, 2009). None of the plants are
known to be hyperaccumulators, even though they have been studied for phytoremediation
purposes. The plants were grown in single culture as well as a polyculture.
32
The pattern obtained through normalisation of the REE to a standard (PAAS or control) was
used as a tool to follow contamination by heavy metals and radionuclides.
Our study shows that metals (REE) are mobilised in a different way by acidic solutions of
different origin, and that organic acids lead to a different fractionation than inorganic ones.
Indeed, the pattern of REE leached by sulphuric acids was qualitatively similar to the one
obtained by water, with much higher amounts leached. A MREE enrichment was noticeable,
and a positive Ce anomaly. It is recording a typical AMD influenced pattern: AMD until a pH
of 4 is characterised by high concentrations in REEs and LREE depletion, Ce-enrichment,
slight MREE enrichment (Grawunder and Merten, 2012; Lei et al., 2008).
In case of leaching with organic acids, the HREE were enriched compared to the LREE. Since
it is not an effect of the pH, it must be due to the specific properties of organic substances
used for leaching, as their complexing properties. This feature was considered as an indicator
for organic substances involved in the leaching of metals.
Bulk soil
Soil Water
Seepage Water SeW
Water
Shoot
Root
Rhizosphere
Plant
Soil
Mass Transfer
Figure 7: Overview over the experimental settings of the pot experiment and the compartments of the studied system.
33
Figure 8: PAAS normalised REE pattern of soil from the test field eluted with water, sulphuric acid, and organic
acids.
The REE patterns of soil, water and plants were compared to controls and to soil eluted with
different solutions as water, inorganic and organic acids in order to define the factors
influencing the metal behaviour.
Figure 9: REE pattern of soil water (in pots with plant growth) relative to soil water of the control pot. A clear
enrichment of HREE compared to LREE is visible in all samples.
SoW: Soil water
A REE fractionation from soil into the soil water and further to roots, and finally from roots to
shoots was observed. The rhizosphere increased locally the soil pH and decreased locally the
amounts of soluble metals in soil and soil water. The characteristic heavy REE enrichment of
the soil water compared to the control soil water was a hint to the influence of organic acids
(cf. Figure 8), although it could not be discriminated if there were no other organics involved,
such as other exudates or microorganisms.
4 THE ROLE OF MICROORGANISMS IN SYMBIOSIS WITH PLANTS IN
THE CONTEXT OF METAL CONTAMINATION The plants and their metal uptake can be influenced by many other factors, since they form an
own microcosm in their rhizosphere. Indeed, many organisms living around the plants
influence their growth and mineral uptake. If more is known about these, they can be used, if
chosen well, to enhance phytoremediation procedures, especially on soil poor in nutrients.
1E-05
1E-04
1E-03
1E-02
1E-01
1E+00
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
/PA
AS
1E-01
1E+00
1E+01
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
SoW
/ So
W c
on
tro
l
Leaching with H2SO4
Leaching with organic
acids
Leaching with water
34
Microorganisms are known to react with metals present in their environment therefore many
are used to treat wastewaters containing high amounts of metals (AMD) by precipitating them
as sulphides, so concentrating or immobilising them, and increasing the pH. These
biogeochemical processes are catalysed mainly by sulphate reducing bacteria like
desulfovibrio or desulfotomaculum (Cohen, 2006). Compost is added to ensure an organic
carbon source for sulphate. Besides there are also Mn-oxidiser and Fe-oxidiser, that
precipitate metal oxides or co-precipitate them as hydroxides. However, even though these
processes are of great interest in liquid media, they play a smaller role in soil. There, on the
other hand, other specific microorganisms interact with plants in the rhizosphere zone. Those
are the interactions on which will be lead the focus on in this study.
Plant roots release a wide range of substances, which, especially the easily decomposed low
molecular weight ones, are involved in attracting beneficial microorganisms and forming
mutualistic associations in the rhizosphere (Marschner, 2012). The most important
mutualisms exist between plants and mycorrhizae or rhizobacteria (Badri et al., 2009), which
can also exist simultaneously and influence each other’s. Relations in nature are more
complex in the rhizosphere and involve a rich and diverse community composed of several
bacteria (endobacteria, pathogens), fungi including Arbuscular Mycorrhizal ones (AM),
micro-fauna (i.e. nematodes), resulting from the specific chemical interactions between the
different organic compounds excreted by plant roots and the different microscopic actors of
the soil. However, the colonisation of roots is not limited to the root surface, but can occur
further inside the root tissues.
4.1 Description of the endophytic population in two plants grown on this
specific contaminated substrate
Bacteria have been known to exist within plant tissue since many years (Tervet and Hollis,
1948); although the pathogenicity was believed to be their main function. Later, some studies
showed that bacteria living within the tissues had no negative and even beneficial effects on
the host (Davison, 1988). From then endophytic bacteria are defined as bacteria residing
within living plant roots without causing substantive damages to their host. While some of
them are very specifically associated to one host, others are more flexible. Root colonisation
is a complex procedure involving several steps (Badri et al., 2009) and establishment in plant
tissue includes several complex mechanisms (Reinhold-Hurek and Hurek, 2011).
Endophytes are distributed in several bacterial phyla, including Firmicutes, Actinobacteria,
Proteobacteria and Bacteroidetes, represented in 82 genera (Lodewyckx et al., 2002). They
are found in various plant tissues, ranging from roots, stems, leaves and seeds (Madmony et
al., 2005; Rajkumar et al., 2009; Reinhold-Hurek and Hurek, 2011). Some are known to be
obligate endophytes and are transmitted over the seeds to the next generation (Majewska-
Sawka and Nakashima, 2004; Mastretta et al., 2009).
Endophytes were found in numerous plant species, and studied mostly in agricultural relevant
plant, although an increased interest is now given to phytoremediation plants, due to the rising
interest to find ways to promote plant survival and growth under these challenging growth
conditions.
35
4.1.1 Identification of the strains
Endophytic bacteria were isolated from the root, stem and leaf of cultivars of Trifolium plants,
and further from shoots and roots of Festuca plants growing on the studied substrate in a pot
experiment. 78 stable, morphologically distinct isolates were obtained belonging to the
several genera of Proteobacteria, Firmicutes, Actinobacteria were found. Many of their most
similar database matches were first isolated from soil and plant, and a relatively big
proportion of them are described in the context of metal contamination or remediation,
especially of organic contaminants. Some endophytic species can fix nitrogen (Sphingomonas
azotifigens isolated from the roots of rice plants, Kamnev et al., 2005), and many are
aromatic-degrading bacteria. The isolates were characterised and tested for properties useful
for plant promotion on metal contaminated soil. So, numerous strains were able to produce
IAA, organic acids, siderophores, and many were resistant to metals as Zn, Ni, Mn Cd or Al
[see Chapter 4]. However, it is important to note that this is a result based on cultivation, and
so non- or not easily cultivable strains are not detected.
4.1.2 Spatial distribution of the strains and bacterial community
The identified endophytic community was different for the two studied plants (Figure 10), so
it seems that a selection takes place. It seems that generally the community is more influenced
by the plant species than the substrate. The main reason is believed to be the active attraction
of bacteria through different organic secretes from the roots, the resulting cocktail being plant
specific, even though influenced by the surrounding conditions [cf. chapters 1&3]. Hence,
each plant species have a different rhizosphere micro-flora in terms of abundance and
physiological characteristics, which can be further modified by the properties of the soil, plant
age and plant nutritional status (Marschner, 2012). Indeed, plants are able to select
specifically from the bacterial pool in the soil which bacterial community would form in their
tissue, probably through the production of different root exudates. (Wang et al., 2008) showed
how the community changed from soil to the roots of different plants, with in particular a shift
from gram positive majority in the soil and gram negative in the plants, and how some genera
were found in grass (Festuca) and not in tree (Betula).
The endophytic bacteria showed clear spatial compartmentalisation within the plant,
suggesting that they can be specific associations with plant tissues and also the possibility of
different uptake mechanisms for different plant tissues.
Figure 10: Diversity assessment of isolated endophytic microorganisms for different compartments (roots R, Stems St and leaves L) of Trifolium pratense (T) and Festuca rubra (F), calculated based on the isolated CFU /g
1E+00
1E+01
1E+02
1E+03
1E+04
1E+05
1E+06
1E+07
1E+08
1E+09
TR TSt TL
CF
U/g
FR FL
fungi
unknown
firmicutes
bacteroidetes
actinobacteria
gamma-proteobacteria
beta-proteobacteria
alpha-proteobacteria
36
4.2 Improving plant growth on heavy metal contaminated soil using
selected endophytic microorganisms
The use of beneficial bacteria to promote plant growth and health has been suggested already
over 20 years ago for agricultural crops (Davison, 1988), and studied later on for microbial
bioremediation of metals and also suggested as inocula to enhance re-vegetation of
contaminated sites, phytoremediation, as reviewed by several authors (Beolchini et al., 2008;
Kidd et al., 2009; Rajkumar et al., 2009; Shetty et al., 1994; Vangronsveld et al., 2009;
Weyens et al., 2009b; Zhuang et al., 2007). Beneficial arbuscular mycorrhiza, yeasts or
various soil bacteria, also called generally PGPR (Plant Growth Promoting Rhizobacteria)
have been used. However, it is crucial to understand more about the plants naturally present in
poor or polluted environments, in order to allow re-vegetation of sites with scarce vegetation
cover, thus being subject to soil leaching and eventually to contamination spreading. In
particular, the development of roots plays a decisive role in the aspect of soil stability.
Vegetation can help to avoid erosion by precipitation and wind and stabilise the soil, reducing
the percolation (Kidd et al., 2009). In the present study, the focus is put on their endophytic
bacteria. A number of isolates demonstrated the ability to produce plant growth promoting
substances and resistance to the present metallic contaminant. These plant fitness-enhancing
properties are of interest for the use of this mutualism for the enhancement of plant fitness and
further for remediation purposes. All these properties are helpful for the survival on metal
contaminated soils. Indeed, growth-promoting properties can be seriously affected by metal
contamination. On the other hand, inoculation of effective N2-fixating strains Rhizobium
leguminosarum bv. trifolii lead to a revival of the nitrogen fixation capacity if the inoculated
cell number was large enough (Giller, 1989). This shows the importance of the good choice of
inocula, especially in specifically contaminated sites.
4.2.1 Plant microorganism partnerships for a better biomass production
Endophytic bacteria can improve plant growth using different processes. Plant associated
bacteria can improve plant nutrition by fixing N2 and solubilising macronutrients as poorly
soluble (P)-minerals, thus delivering nutrients normally unavailable for plants and fulfil so the
role of natural fertilisers (Badri et al., 2009; Weyens et al., 2009a; Yanni et al., 1997).
Nitrogen fixers were found in Archea (Methanosarcina) and many bacterial genera, mainly
proteobacteria as Sphingomonas (S. azotifigens), Burkholderia, Pseudomonas, Azotobacter,
competing ions, complexing agents Adsorption increases with the pH and the content of
organic matter, decreases with competing cations or
dissolved ligands
Pb Carbonate in the soil
Cu Organic matter (strong) / Fe- and Mn-oxides /sulphides / carbonates
Cu exist in solution mainly as complex with soluble organic substances
Mn Clay minerals Adsorption stronger with increasing pH
Stronger retention by carbonate containing soils
(Precipitation as MnCO3)
Zn Clay minerals Non-available in calcareous and alkaline soils because
of sorption on carbonates as Zn-oxide, Zn-carbonate or
Ca-zincate
Co Fe- and Mn-Oxides
Since the speciation of metals is essential for their behaviour in the soil-water system and for
their interaction with the biosphere, different methods have been developed to estimate the
proportion of metals that are soluble, and those bound to different phases of the soil, more or
less easily released. Commonly, a sequential extraction is used. It means that soil samples are
treated with solutions of increasing extraction potential. Different methods of simple or
sequential extraction have been proposed and compared (Beolchini et al., 2008; Doelsch et al.,
2008; Krasnodebska-Ostrega et al., 2009; Lewandowski J., 1997; Ma and Uren, 1997; Martin
et al., 1987; Sauerbeck and Lübben, 1991), to describe this solubilisation and repartition of
heavy metals, and the one according to Zeien and Brümmer (1989) has been mainly used for
the present soil (Figure 2).
57
ROOT SOIL
INTERFACE
Mass Flow
Soil solution
Cation exchange
Organic Chelators
unsoluble
Erosion
M-Oxides
Plant rests
Diffusion M+ M+
M+
H+
M+
M+
M+
M+ M+
M+
Ca+ or H+
M+
M+
M+
M+
M
M+ Element Cation
M+ Element Cation - complexed
M Element (Nutrient)
Cd Mn Co Pb Cu Ni Fe Zn Al
Mobile
Exchangeable
Mn-Oxides
Organics
amorphous Fe-Oxides
crystalline Fe-Oxides
Residual fraction
mainly occurring as
partly or in certain soil types occurring as
Figure 2: Main fraction according to Zeien and Brümmer’s sequential extraction, in which metals are mainly found to bind to (Paas, 1997).
However, despite the detailed description of binding phases for metals, the active
influence of plants by their root exudates was not taken enough in consideration. One
question rising concerning the availability of metals in the soil (Allen and Janssen, 2006),
especially regarding the sequential extraction is how exudation influences the fraction of
easily extractable metals. In other words, the available fraction could be higher than the easily
dissolved one found following the classical sequential extraction method. In particular
Puschenreiter et al. (2005) suggest that root activities, such as the exudation of organic acids,
triggered the replenishment of soluble metal from immobile metal fractions of the soil.
Another example is given about the bioavailability of Pb near the roots of rice plants (Lin et
al., 2004). It is well known that the organic acids produced by roots can complex metals -
Díaz-Barrientos et al. (1999) compare them in their study about sequential extraction with the
properties of EDTA - and so modify their availability. This has been noticed by different
authors and is an invitation to consider this aspect when estimating bioavailability of metals in
the environment (Haoliang et al., 2007; Mucha et al., 2005).
Figure 3: Important abiotic processes for the uptake of trace elements (M+). Trace elements present in form free ions or
soluble chelates can be taken up by plant roots. Metals bound to soil particles or present as insoluble compounds can be
mobilised by the action of root exudates, i.e. acidic substances which dissolve metallic compounds, by cation exchange, or by secretion of chelating components. (after Mitchell, 1972)
58
Important factors influencing the solubilisation of metals by
plants through the quantity and composition of root exudates:
(5) root-induced changes in pH of the rhizosphere
(6) complexing capacity of organic compounds released
(7) reducing capacity of the roots
(8) need for nutrients in particular essential trace elements
2 ROOT EXUDATES: THE ACTIVE CONTRIBUTION OF PLANTS TO
METAL MOBILITY
Availability of trace elements to plants is governed by the dynamic equilibrium between
aqueous and solid soil phases, rather than by the total metal content. To satisfy physiological
needs for nutrients or to avoid metal toxicity, plants are able to modify clearly the mobility of
metals.
In the soil solution, elements are present as free ions, ion pairs, ions complexed with organic
anions, and ions complexed with organic macromolecules and inorganic colloids. The most
important metal pools in the solid phase include the exchange complex, metals complexed by
organic matter, sorbed onto or occluded within oxides and clay minerals, co-precipitated with
secondary minerals (e.g. Al-, Fe-, Mn-oxides, carbonates and phosphates, sulphides) or as part
of the crystal lattices of primary minerals (Kidd et al., 2009). Not all metal ions necessarily
occur as cations; for example many elements occur as oxyanions like arsenate, selenate,
selenite, chromate, chromite, which is also important among others for their behaviour in
biological systems (Clarkson, 1993).
Plant-induced modifications of trace element speciation and bioavailability in the rhizosphere
are the result of sharp biogeochemical gradients in elemental concentrations, pH, pCO2, pO2,
redox potential and organic ligand concentrations, and microbial biomass (Dakora and
Phillips, 2002; Kidd et al., 2009). The potential changes depend on the air content of the soil;
the roots can influence this values by excreting different substances as acids, protons or
chelators, and so influence the availability and uptake of mineral nutriments (Marschner,
2005). Although plants are able to influence their environment by acidification with protons
or by secreting CO2 (Dakora and Phillips, 2002), the main influencing factor is given by the
secretion of diverse organic components.
Rhizodeposits include a wide spectrum of components ranging from simple chemical exudate
compounds to entire root fragments, originating from dead cells. They can be grouped into
five general classes: exudates (amino acids, low-molecular-weight carboxylic acids, sugars,
and simple and flavonoid-type phenolics), secretions , plant mucilages, mucigel, and root
lysates (Curl and Truelove, 1986 in (Kidd et al., 2009). Among the different substances
secreted by plants,
organic acids compose 1-
3% of the dissolved
organic carbon (DOC) in
the soil. Root exudates
and dead root material
may comprise 30-40% of
the total organic matter
input to soils. This is
released into the rhizosphere, which constitutes only 2-3% of the total soil volume (Grayston
et al., 1996).
Root exudates play a role in the weathering of soil, for the mobilisation of nutrients as P,
NH4+ or Fe, especially through the action of organic acids, phytosiderophores, phenolic
compounds. They are further important for the protection of plants against uptake of heavy
59
metals into the roots - mainly citrate, malate, small peptides are important in this case, or for
attraction of beneficial microorganisms through phenolic compounds, organic acids, sugars
(Grayston et al., 1996; López-Bucio et al., 2000; van Hees et al., 2003). Root exudates can
also act as signal molecules or as precursors for hormones. Very low concentrations are
sufficient for their biological effect (Marschner, 2005).
Table 3: Chemical classes of roots secretes (Modified from Bolton et al. 1992)
organic acids by a marsh plant and implications on trace metal availability in the rhizosphere of estuarine sediments. Estuar. Coast. Shelf S. 65, 191-198.
Nahas, E., 1996. Factors determining rock phosphate solubilization by microorganisms
isolated from soil. World J. Microb. Biot. 12, 567-572.
Natarajan, K.A., 2008. Microbial aspects of acid mine drainage and its bioremediation. T.
Nonferr. Metal Soc. 18, 1352-1360.
Neagoe, A., Merten, D., Iordache, V., Büchel, G., 2009. The effect of bioremediation methods
involving different degrees of soil disturbance on the export of metals by leaching and by
plant uptake. Chem. Erde - Geochem. 69, 57-73.
Oburger, E., Kirk, G.J.D., Wenzel, W.W., Puschenreiter, M., Jones, D.L., 2009. Interactive
effects of organic acids in the rhizosphere. Soil Biol. Biochem. 41, 449-457.
Paas, N., 1997. Untersuchungen zur Ermittlung der geochemischen Barriere von Gesteinen
aus dem Umfeld untertägiger Versatzräume im Steinkohlenbergbau des Ruhrkarbons. Rheinische Friedrich-Wilhelms-Universität, Bonn, p. 234.
Pathak, A., Dastidar, M.G., Seekrishnan, T.R., 2009. Bioleaching of heavy metals from
sewage sludge by indigenous iron oxidizing microorganisms using ammonium ferrous sulfate
and ferrous sulfate as energy source: a comparative study. J. Hazard. Mater.
Pollmann, K., Raff, J., Merroun, M., Fahmy, K., Selenska-Pobell, S., 2006. Metal binding by
bacteria from uranium mining waste piles and its technological applications. Biotechnol. Adv.
24, 58- 68.
Pulford, I.D., Watson, C., 2003. Phytoremediation of heavy metal-contaminated land by trees
- a review. Environ. Int. 29, 529-540.
Puschenreiter, M., Schnepf, A., Millán, I.M., Fitz, W.J., Horak, O., Klepp, J., Schrefl, T.,
Lombi, E., Wenzel, W.W., 2005. Changes of Ni biogeochemistry in the rhizosphere of the
exudation of protons and citrate in Lupinus albus as affected by localized supply of phosphorus in a split-root system. Plant Sci. 168, 837-845.
Shen, W.-b., Yang, H.-q., 2008. Effects of earthworm and microbe on soil nutrinients and
heavy metals. Agr. Sci. China 7, 599-605.
Sheng, X.-F., Xia, J.-J., Jiang, C.-Y., He, L.-Y., Qian, M., 2008. Characterization of heavy
metal-resistant endophytic bacteria from rape (Brassica napus) roots and their potential in promoting the growth and lead accumulation of rape. Environ. Pol. 156, 1164-1170.
Shetty, K.G., Hetrick, B.A.D., Figge, D.A.H., Schwab, A.P., 1994. Effects of mycorrhizae and
other soil microbes on revegetation of heavy metal contaminated mine spoil. Environ. Pol. 86,
181-188.
Sipos, P., Póka, T., 2002. Threshold limit values for heavy metals in soils in the function of
spacial and temporal variation of geochemical factors, XVII. Congress of CBGA. Special
Issue of Geologica Carpathica (CD), Bratislava.
Sreekrishnan, T.R., Tyagi, R.D., 1995. Sensitivity of metal-bioleaching operation to process
amongst organic matter content, heavy metal concentrations, earthworm activity, and soil microfabric on a sewage sludge disposal site Geoderma 57, 89-103.
Usman, A.R.A., Mohamed, H.M., 2009. Effect of microbial inoculation and EDTA on the
uptake and translocation of heavy metal by corn and sunflower. Chemosphere 76, 893-899.
Zhuang, X., Chen, J., Shim, H., Bai, Z., 2007. New advances in plant growth-promoting
rhizobacteria for bioremediation. Environ. Int. 33, 406-413.
81
Chapter 2
STUDY OF THE INFLUENCE OF PLANT ROOT EXUDATES ON
HEAVY METAL MOBILITY BY MEANS OF ANALYSIS OF RARE
EARTH ELEMENT FRACTIONATION PATTERNS
Tsilla Boisselet* , Dirk Merten, Georg Büchel
1 Institute for Geosciences, Friedrich Schiller University, Burgweg 11, 07743 Jena, Germany
The present chapter is the manuscript of an article to be submitted.
82
Chapter 2: Study of the influence of plant root exudates on heavy
metal mobility by means of analysis of rare earth element
fractionation patterns
ABSTRACT
Root exudates play a key role for the bioavailability of trace metals and their toxic effects on
the biosphere. A pot experiment was performed in order to study the effect of plants on
metal mobilization and uptake on a soil from a former uranium mining site contaminated by
several trace metals, overall Mn, Al, Ni, Zn, U and Rare Earth Element (REE). Plants used
were clover (Trifolium pratense), red fescue (Festuca rubra), sunflower (Helianthus
annuus) and Triticale. The REE patterns of soil, water and plants were compared to controls
and to soil eluted with different solutions as water, inorganic and organic acids in order to
define the factors influencing the metal behaviour. It was observed that a heavy REE
(HREE) enrichment compared to MREE (middle REE) and LREE (light REE) occurred in
the solution only after leaching soil with organic acids. The root REE were 10-fold
concentrated and showed a fractionation from soil solution to the roots. A consistent LREE
enrichment of the shoots relatively to the roots was observed for all plants. For sunflower
also MREE enrichment was present.
After a balance calculation, simultaneously to an increased locally the soil pH, a local
decrease of the amounts of soluble metals in soil and soil water was observed in the
rhizosphere. The characteristic HREE enrichment compared to previous elution
experiments, of the soil water compared to the control soil water was a hint to the influence
of organic acids, although some further factors could be involved too, such as other exudates
or microorganisms.
1 INTRODUCTION Many different factors have an influence on the mobilisation and immobilisation of metals in
soil and water. These are mainly soil pH and oxygen content, redox conditions, soil
composition allowing cation exchange or precipitation, soil weathering as well as the type of
metal and its chemical form, or heavy metal competition (Blume and Brümmer, 1991; Bradl,
2004; Karim and Khan, 2001; Marschner, 2005; Paas, 1997). Additionally, microorganisms
and plants are able to change these physicochemical parameters, and provoke themselves
further mineralisation or mobilisation reactions, by acidification of the soil through exudates,
active redox reactions for energy production, precipitation of minerals and complexation of
metals, among others. The investigation of the speciation of metals and its changes are
essential for risk assessment, toxicity studies or remediation techniques (Díaz-Barrientos et
al., 1999; Kidd et al., 2009).
Among the different substances secreted by plants, organic acids compose 1-3% of the
dissolved organic carbon (DOC) in the soil. Root exudates and dead root material is supposed
to comprise 30-40% of the total organic matter input to soils (Grayston et al., 1996). They are
released into the rhizosphere, which consists of the root-influenced soil volume up to few
millimetre from the root surface, and constitutes only 2-3% of the total soil volume (Grayston
83
et al., 1996). Root exudates play a role in the weathering of soil, for the mobilisation of
nutrients as phosphorus or iron; especially organic acids, phytosiderophores, phenolic
compounds are involved in these processes. Further they are important for the protection of
plants against uptake of heavy metals into the roots through the action of citrate, malate, small
peptides, or for attraction of useful microorganisms by phenolic compounds, organic acids,
sugars (Grayston et al., 1996; López-Bucio et al., 2000; van Hees et al., 2003). Root exudates
can also act as signal molecules or as precursors for hormones; very low concentrations are
sufficient for their biological effect (Marschner, 2005). In the past, some methods have been
used to quantify the metals bound to different fractions of the soil, especially the sequential
extraction (Zeien and Brümmer, 1989). In particular, ammonium nitrate extraction was found
to be a good estimation for the plant available fraction and used as a DIN norm in soil
analysis. However, on one hand this estimation was not valid for all soil and all metals
(Gryschko et al., 2004), and on the other side the amount and the proportion of the different
acids in the soil depend on the plant species, the cultivation time, the constitution of the soil,
as well as the age of the plant and the distance to the root (Grayston et al., 1996).
Furthermore, huge differences in pH and metal mobility are possible between rhizosphere and
bulk soil. These processes can strongly influence the mobility of heavy metals in the
surrounding soil. To study the effect of the rhizosphere on metal mobility a pot experiment
was designed.
The soil originated from the Ronneburg mining district in Thuringia (Germany) which was a
large uranium producing area until 1990 (Jakubick et al., 2002; Kahlert, 1992; Lange, 1995).
The mining activities strongly altered the hydrogeology of the area. During mining,
exhumation of sulphide minerals as well as acid-leaching (10g/L sulphuric acid) of waste
heap led to metal dissolution, due to pyrite oxidation. Later, flooding and precipitation formed
so-called acid mine drainage that infiltrated into the soil. These acidic and highly mineralised
solutions infiltrated into the soil, and polluted the water-soil system with high concentrations
of uranium, Rare Earth Elements (REE) and other metals. In the 1990s the heap and 10 m of
underlying Quaternary sediments were filled into the nearby open pit Lichtenberg and the
basement area was remediated. Despite remediation activity, contamination is still
measurable.
Since the soil pH is quite low (pH 4-4.5), the mobility and bioavailability of metals is likely to
be high and so the up-taken amounts of metals by plants would be significant, as monitored
before. Therefore, this area is adequate for monitoring of groundwater chemistry and soil
parameter and improving remediation strategies for slightly heavy metals contaminated areas.
Four different plants were used in the present study. In order to reconstruct the natural
conditions and understand on-going processes in the study area, two autochthones plants were
chosen. Festuca rubra is a very resistant grass found in many heavy metal contaminated
areas, and known as a pioneer plant. Clover was chosen because of its ability to fix air
nitrogen due to its symbiosis with bacteria, and so to overcome partially the nitrogen-poorness
of the soil. Additionally, reference plants (sunflower and Triticale) enable comparison with
previous studies, especially regarding REE fractionation (Kidd et al., 2009; Lonschinski,
2009). None of the plants is known to be hyperaccumulators, even though they have been
studied for phytoremediation purposes, mostly for phytostabilisation. The plants were grown
in single culture as well as a mixed culture.
84
REE are elements of the lanthanide group (La-Lu), including in some cases also Y and Sc.
These metals are very similar to each other because of their similarity in the electron
structure. They are strongly electropositive, and occur mostly in oxidation state 3. They are
considered as chemical analogues to trivalent actinides (Ding et al., 2006). Only Ce und Eu
can be found in other valences (Ce+4
or Eu+2
), which causes a different chemical behaviour
depending on the redox potential. Rare Earths Elements are often separated in light (LREE),
middle (MREE) and heavy REE (HREE). The pattern obtained through normalisation of the
REE with a standard (in this case PAAS, Post Archean Australian Shale, (McLennan, 1989) is
a tool to study water-rock-interactions, as tracer for erosion processes, or to follow the flow of
water and in case of AMD influenced areas to follow contamination by heavy metals or
radionuclides (Aubert et al., 2001; Merten et al., 2005). Possible host phases of REE in nature
are Fe-hydroxides, Mn oxides, clay minerals, and humic substances. Different factors can
influence the pattern of REE. Not only the source material causes a typical pattern, but also
pH (precipitation, sorption and desorption, dissolution), redox-conditions and ligands can alter
it. In particular, organic substances excreted by plants to optimize their nutrient input can play
a role in the regulation of the soil pH and as well complex different metals (Pourret et al.,
2008; van Hees et al., 2003).
This study aims at investigating the influence of rhizosphere processes on metal mobility,
using REE pattern as a tool. Therefore, four different plants (clover and red fescue as
autochthonous plants and sunflower and triticale as model plants, as monocultures and
polyculture) were grown on contaminated substrate, and the total metal contents of soil, soil
water, seepage water and plants, as well as the soluble metal content of the soil in the
rhizosphere and in the bulk soil were measured. The REE patterns of all compartments were
compared to those obtained by leaching the soil with different reagents, in order to find an
explanation for the mechanisms involved.
Indeed, it is important to understand how plants and their rhizosphere influence metal
mobility, in particular to estimate the bioavailable fraction beyond sequential extraction, in
order to optimize and use these processes for phytoremediation purposes.
85
Bulk soil
Soil Water
Seepage Water SeW
Water
Shoot
Root
Rhizosphere
Plant
Soil
Mass
Transfer
Analytical parameters Plant (Shoot + Root) Biomass [g] Total metal content [µg/g]
Soil (Rhizosphere + Bulk)
Elution – soluble metal content [µg/g] Water content [%] Total metal content [µg/g] pH
Electrical conductivity
Water (Soil Water + Seepage Water)
Metal content [µg/L] pH
Electrical conductivity
Filter
Collection pot
Pot
Sampler (suction tube) with
taping point for syringe
Drainage
Foil
Figure 1: Overview over the experimental settings of the pot experiment and the compartments of the studied system, including the analytical parameters
86
2 MATERIALS AND METHODS
2.1 Substrate
The substrate sampled at the remediated northern part of the base area of the former leaching
heap called Gessenhalde at the former Uranium mining site. The soil that was used to test
remediation strategies was homogenized on the top 100 cm. It was air dried and sieved to a
grain size of < 2 mm. It consists in silty sand (Mirgorodsky et al., 2010) with a high content in
clay minerals. The soil pH is quite low (pH 4-4.5). The nutritive quality of the soil is weak, at
least concerning the available part of nutrients. So, the contents of organic carbon and
inorganic nitrogen are very low (Table 1) compared to an optimal soil as Rendzina soil.
Phosphorus is also present in low amounts, most of it being insoluble. Similarly, Fe is poorly
present in a plant available form, most of it being found as iron oxides, and in the residential
fraction (Table 2). Mg is the only macroelement present in sufficient amounts, whereas S is
present in high amounts. The soil was furthermore characterised by a moderate contamination
with metals (Table 2) including REE (La-Lu), with average total amounts of ∑REE of about
180 µg/g. Other important contaminants are Al, Ni, Zn and Cu, and remaining U from the
mining.
Table 1: Substrate characterisation and comparison with a soil adapted to plant growth (Mirgorodsky et al., 2010)
However, in all experiments the electrical conductivity (130 - 420 µS/cm) was inversely
correlated to the pH, which shows that pH is a main factor determining the amounts of
dissolved elements in soil and water.
The amounts of metals collected through leaching with water are very low and in some cases
below detection limit, except for the pots without plants and at less extend those planted with
sunflower. Despite this, it was still possible to notice a trend: the electrical conductivity was
lower in the rhizosphere compared to controls or bulk soil, showing that the amounts of
soluble metals were lower.
Essential plant nutrients as Mg, K, Na and S are present in the water-soluble fraction of the
soil (Table 5). The amounts of K and Na are not strongly affected by plant growth, only K is
found in higher amounts in the soluble fraction of the rhizosphere soil of sunflower. S and Mg
are found in high amounts (over 5 mg S/g) in the soluble bulk soil soluble fraction of pots
planted with sunflower, Trifolium and plant combination. Phosphorus is below detection limit
in the water-soluble fraction of the soil (Table 5). Manganese content is less than 100 µg/g in
the water-soluble fraction for control and most of the samples, only Trifolium and the plant
combination show amounts between 250 and 600 µg/g. The higher contents are found in the
bulk soil (for instance Trifolium bulk soil has 420±12 µg/g Mn in the soluble fraction, the
corresponding rhizosphere 235±31 µg/g). This trend is found for many other metals such as
Al (about 2 µg/g vs. 5-15 µg/g for Trifolium and plant mixture), Cd, Zn and Co. For Ni and
Cu it was similar, except that the rhizosphere of all plants and Festuca soil had lower amounts
than unplanted soil. Values for Fe were mostly under detection limit or very low. The
decrease of soluble metal content in the rhizosphere soil was in particular true for REE, which
were used as a tracer to explain the metal uptake and show the active action of plant exudates
(Table 5).
The amounts of REE for each sample of the elutions were normalized to the PAAS standard
(Post Archean Australian Shale) (McLennan, 1989) and are plotted in Figure. The REE
patterns show a slight positive Ce anomaly, except for the water elution (see Table 6, Figure
3), and an enrichment in MREE. Generally, LREE are depleted compared to HREE. The
pattern is not much different comparing elution with water and elution with sulphuric acid
(pH 0.9). It could be observed that the decrease of pH of the solution increased the amounts of
leached REE: for sulphuric acid leaching (see Figure 3) the leached amounts are at least one
order of magnitude higher than for the other solutions. The presence of organic acids
compared to inorganic acids lead to a fractionation of the REE present in the soil, resulting in
a significant HREE enrichment (see also Figure 3). Indeed, Lu/La values are much higher for
organic leaching (18.9 ± 14) than for other leaching with water or sulphuric acid (respectively
1.6 ± 0 and 2.0 ± 0.4).
Table 6: Ce anomalies and HREE enrichment of patterns after leaching the studied substrate with three different solutions
Ce/Ce* (1) SD Ce/Ce* (2) SD Lu/La SD
H2O 0.25 ±0.97 0.87 ±0.14 1.57 _
Org 1.64 ±0.20 1.40 ±0.23 18.94 ±14.41
H2SO4 1.36 ±0.01 1.37 ±0.05 1.96 ±0.42
94
Figure 3: PAAS normalized REE pattern of soil from the test field eluted with water [dash lines], sulphuric acid
[point-dash lines], organic acids [plain lines].
If REE patterns are used regarding to this result, a different pattern in soil water sampled near
by the roots compared to seepage water can be expected.
Considering the water phase, the REE patterns normalized to PAAS do not show a visible
HREE enrichment. On the other hand, if the water is normalized to the control pot, the effect
is much more visible (Figure 4). The pattern of the soil waters normalised to the soil water
from control pot (Figure 4) has this enrichment observed after eluting soil material with
organic acids; this could be a hint for the influence of present organic acids in the root zone.
Figure 5 shows that the HREE enrichment in the soil water corresponds to a HREE depletion
in the soil total amounts.
REE can also be traced within the plant from the roots to the shoots. We could observe that
here also REE behave in a different way.
Figure 4: REE pattern of soil water (in pots with plant growth) relative to soil water of the control pot. A clear enrichment of HREE compared to LREE is visible in all samples.
1E-05
1E-04
1E-03
1E-02
1E-01
1E+00
La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
/PA
AS
1E-01
1E+00
1E+01
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
SoW
/ So
W c
on
tro
l
95
Figure 5: REE pattern of soil total contents of sample relative to control on the example of Trifolium: it is clear that the depletion of HREE is located in the rhizosphere. Diamonds stand for the rhizosphere, full line for bulk soil.
The amounts of REE are 10 fold higher for the roots than for the shoots, except for clover
where the ratio between shoot and roots is smaller (Table). It is also noticeable that the total
REE amounts per g dry biomass are lower at mixed crop cultivation than at single crop
cultivation (Figure 7, Table 7), except for Festuca, which is taking up more metals as
polyculture. Further the differences between patterns are more marked between shoots and
roots than between plants. The REE patterns are similar for all plants when normalized to
PAAS (see Figure 7): MREE enrichment and a slight positive Ce anomaly (Ce/Ce* = 1.2-
1.4); the LREE to HREE ratio is variable (Lu/La 0.7-3.4). It was also observed that generally
HREE are less translocated than LREE. The same was observed for many metals, as Al, Fe,
and most of the contaminants. Only Mn had a translocation factor of over 1, and additionally
Cr and Zn in the case of Trifolium (results not shown).
Table 7: Translocation factor for REE from root to shoots ; TF=metal content in shoots/metal content in roots
TF La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Monoculture
To visualise the REE fractionation from the environment to the plants, the REE amounts of
the roots are normalised to the soil water (Figure 8), since it is assumed to be the fraction
taken up by the plants. The pattern shows a fractionation in a wavy shape: elements from Pr to
Eu on one hand and Yb and Lu on the other hand are clearly enriched.
For the fractionation within the plant, shoots total digestion are normalised to root total
digestion (Figure 9). There is little fractionation from shoots to roots, though a clear LREE
enrichment is observable (La/Lu ranging from 1.3 to 3), except for triticale. The patterns look
similar to those shown by (Lonschinski, 2009) with similar plants on a similar substrate. In
that study, sunflowers show a clear depletion of HREE to LREE, with a slight MREE
1E-01
1E+00
1E+01
La Ce Pr NdPmSm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
to
tal
sam
ple
/ su
bst
rat
Trifolium B
Trifolium Rh
96
enrichment, and a slight positive Gd anomaly. Festuca rubra showed an almost continuous
depletion in REE from light to heavy.
Figure 6: PAAS-normalized REE patterns of the soil water (SoW) and seepage water (SeeW) for pots planted with
sunflower
Figure 7: PAAS normalized REE patterns of all plants (roots and shoots, total digestion) normalized to PAAS. [a] Sunflower, [b] Festuca, [c] Trifolium, [d] Triticale
1E-03
1E-02
1E-01
1E+00
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
/PA
AS
Control SoW
Control SoW
Control SeeW
Control SeeW
Helianthus SoW
Helianthus SoW
Helianthus SoW
Helianthus SeeW
Helianthus SeeW
1E-03
1E-02
1E-01
1E+00
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
/ P
AA
S
1E-03
1E-02
1E-01
1E+00
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
/ P
AA
S
1E-03
1E-02
1E-01
1E+00
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
/ P
AA
S
1E-03
1E-02
1E-01
1E+00
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
REE
/ P
AA
S
[a] [b]
[c] [d]
97
Figure 8: Plant REE patterns normalized on the corresponding soil water: fractionation from soil water to plant root
Figure 9: REE patterns of shoots (total digestion) standardized on roots
Stars stand for sunflower, circles for triticale, triangles for Festuca rubra and squares for clover. Plain symbols are for monoculture, empty ones for mixed crops.
4 DISCUSSION Plants grown on the contaminated soil poor in nutrients reacted in different ways to their
environment depending on the plant species and on if they were grown as monocultures or in
combination with each other’s. Their biomass production was affected by the soil conditions,
and vice versa the plants influenced the mobility of elements in their rhizosphere soil and
water, by excreting organic substances and changing the pH. In particular, REE were
dissolved, taken up and fractionated on their way from the soil solid phase over the soluble
phase in the soil water to the plant roots and shoots. This particular fractionation is a clue for
the specific action of plant roots in the soil.
4.1 Plant growth and biomass production There were strong differences in the biomass of different plants. Clover grew very well,
possibly due to its capacity to use nitrogen from the air in association with N2-fixing bacteria.
The biomass of other plants was much lower. They seemed to try to produce seeds as fast as
possible (Wierzbicka and Panufnik, 1998) and showed much reduced growth, and also
unhealthy colour in general, as the symptom of the conjugated effect of lack of some nutrients
1E+00
1E+01
1E+02
La Ce Pr NdPmSm Eu Gd Tb Dy Ho Er Tm Yb Lu
Ro
ot
REE
/ S
oW
REE
1E-02
1E-01
1E+00
La Ce Pr
Nd
Pm Sm Eu Gd Tb Dy
Ho Er Tm Yb Lu
REE
sh
oo
t/ R
EE r
oo
t
Helianthus
Triticale
Festuca
Trifolium (poly)
Festuca (poly)
Helianthus
(poly)
Helianthus
Triticale
Festuca
98
and toxic doses of some metals. Clover did not show signs of disease until at least nine weeks
after germination, where diverse parasites as fungi colonized the leaves. It could be that the
poor soil but probably more the heavy metals present affected the resistance of the plant.
4.2 Plant influence on element mobility in soil and soil water Plant growth influences soil pH, one important soil parameter for metal mobility and
bioavailability. In the present study, the pH was more acidic in non-planted soil than in the
rhizosphere soil. This trend was not expected, since the production of acids should decrease
the pH. However, the production of organic acids was not detected, and other organic
components may result in the same REE pattern without decreasing the pH, for instance by
complexing effects. Since organic acids have a short life time in the rhizosphere - 6 to 12 h
according to Marschner (2012), it could be that there were degraded despite of the precautions
taken during sampling. The amounts should be locally, if a ratio solid to liquid of 1:10 is
given, in the order of magnitude of 10 mg/L or more in order to be able to influence the REE
pattern according to some previous elution experiments with organic acids. Other studies
suggested that under a pH value of 5.5, to avoid possible toxic effects by high metal mobility,
plants would stop producing acids (Dakora and Phillips, 2002). It is to remember that metals
can be mobilized actively by plants, not only through the action of excreted organic acids, but
also by the release of organic chelating agents as siderophores (Römheld and Marschner,
1986). These do not influence the pH of the medium.
It was particularly noticeable for Trifolium and in the polyculture. The dense root net of
Festuca did not allow the soil to be separated into bulk and rhizosphere, therefore all soil was
considered as rhizosphere. Indeed, the pH of soil planted with Festuca showed a higher pH.
On the contrary, no significant pH differences were observed in the case of triticale. This
could be due to the fact that there was very little soil adhering to the roots, which were poorly
developed, so that the rhizosphere soil consisted in realty of a much smaller volume as really
sampled.
The leaching with water showed the highest electrical conductivity and amounts of leached
metals in the control samples. The mobile fraction is in fact more easily taken up by the plants
and it is likely to be the first to decrease locally after plant growth, even if it is possible that
root exudates mobilise less mobile fractions. The same difference could be observed between
bulk soil and rhizosphere: there are less easily available metals in the rhizosphere. Among all
plants, Triticale seemed to have to least effect on soil. This is quite expectable, since its root
system was poorly developed. Trifolium had a big influence on the amounts of soluble metals
in soil, and this effect was also visible and even stronger when it was planted together with
other plants. Festuca, generally known to be an excluder plant, showed the lowest amounts in
soluble metals; it could be that they become immobile in the rhizosphere of the plant; in fact,
the amounts of soluble elements were equivalent or lower than those found in the control. So,
for instance Al, Zn and especially Ni and Cd were found in lower amounts in the soluble
fraction of the soil influenced by Festuca than in the control. Generally, the plant consortium
showed higher amounts of soluble metals than monocultures. Phosphorus is not detectable in
the water soluble fraction of the soil. Though, some is found in the leaves and roots. This
should be due to some active mechanism of the plants, which can mobilize P from less mobile
fractions of the soil. One possibility is that some acids mobilise P from the soil, as suggested
by Cheng et al. (2004) and Shen et al. (2005). However, we found a slightly elevated pH in
99
the rhizosphere, so that an enzyme activity of phosphatase by the plant or possible associated
microorganisms seems to be more probable (van Aarle and Plassard, 2010).
Even though depletion of labile (i.e. the easily dissolved part) metal pools in the rhizosphere
of hyperaccumulator plants often has been found to be associated with sustained or even
enhanced solubility (i.e. soil solution concentration) direct evidence for mobilization of
metals, either due to acidification (Bernal et al., 1994; Li et al., 2003; McGrath et al., 1997) or
induced by root exudation (Salt et al., 2000; Zhao et al., 2001) has not been really put in
evidence so far. Nevertheless, the decrease of Ni in the rhizosphere of the hyperaccumulator
plant Thaspi goesingenseis for instance was clearly related to excessive Ni uptake and
consistent with previous field observations (Wenzel et al., 2003). The interactions of organic
acids released by roots with the soil solid phase appeared to be among the key processes
(Puschenreiter et al., 2005). In particular, the authors suggest that root activities of
accumulators such as the exudation of organic acids triggered the replenishment of soluble Ni
from immobile metal fractions of the soil.
4.3 REE fractionation and specificity of the HREE enrichment The use of the REE fractionation could lead to an explanation for the on-going processes.
However, many processes can modify REE patterns. Generally, the chemical behaviour of
REE is known to be strongly related to the one of Al and Fe especially as Al- and /or
Fe(hydr)oxides. In acid environment, Ln (Lanthanides) occur as LnSO4+ or as Ln
3+. If there
are no ligands in the solution, that keep HREE dissolved, they are mostly bound by oxy-
hydroxides (Aström, 2001). Indeed, in the present case REE seem to follow the same trend as
Al, being depleted in the soluble fraction in the rhizosphere. The REE patterns of the
rhizosphere soil compared to bulk soil show a HREE enrichment. The same is for the soil
water sampled near the roots if normalised to the control pot. There are different processes
that can cause this fractionation: for instance, Fe-oxyhydroxides precipitating scavenge
preferably LREE (Steinmann and Stille, 2008) and result in a relative HREE enrichment
compared to LREE in the remaining Al and Fe oxide particle in the colloidal fraction of the
soil solution. In particular, compared to the other REE, the elements La, Gd, and possibly Lu
show a significantly lower affinity for the Fe-oxy-hydroxides (Bau, 1999). HREE on the other
side tend to bind to Al (Lei et al., 2008). If we relate this to the leaching experiment with
organic acid mixture, and given that that this enrichment is not present in bulk soil if
normalised to PAAS, we can assume that this is an indication for the presence of organic
acids in the close proximity of the roots. HREE enrichment can furthermore come from
binding to carbonates (Pourret et al., 2008). In our case though, since the pH is acidic, it is not
likely that this is the reason for the HREE enrichment. Further, Ce (in case of alkaline water)
and LREE bind preferentially to humic acids, the more alkaline the medium the stronger the
fractionation. The presence of organic acids as citric, malic, tartaric acid increased desorption
of REE. High concentrations of humic acids in the solution increases the adsorption of LREE
on kaolin (Wan and Liu, 2006). This results in a negative Ce anomaly and HREE enrichment
in the liquid dissolved phase. Nevertheless, this effect is limited and probably not visible in an
acid environment. In other studies, it was found that HREE enrichment in biofilms in natural
waters is possibly due to the presence of phosphate sites, and that generally HREE
enrichment can possibly be found in any phases containing phosphate sites (Takahashi et al.,
100
2010). However, on the present site phosphorus plays a very little role and so cannot explain
the observed HREE enrichment in the soil.
Therefore it is reasonable to connect the HREE enrichment to an effect of organic
components excreted by plant roots. In effect, adsorption by organic substrates can produce
heavy REE enrichments in water relative to the LREE, Gd solution enrichments relative to
Eu, and, at relatively low carbonate ion concentrations, enrichments of very light REE
compared to their immediate neighbours according to the work of (Stanley and Byrne, 1990).
Stern et al. (2007) further reported that REE binding to humic substances may display a
regular increase from La to Lu. This enrichment can therefore be considered as characteristic
for mobilisation of REE by organic acids or other organic components not be detected by ion
chromatography, since other parameters as soil composition are equal, and the decrease of pH
between water and very acid sulphuric acid did not change in this way the qualitative
appearance of the pattern. As a consequence, the signature given by the REE shows the
importance of some organic components present in the rhizosphere even though these cannot
be detected anymore, thereby overcoming the analytical short comes.
4.4 Metal uptake and REE patterns in plants The major part of the REE which are taken up by the plant remains in the roots, only 10% is
translocated to the shoots, which is expectable for non-accumulator plants. It was also
observed that generally HREE are less translocated than LREE. The same was observed for
many metals, as Al, Fe, and most of the contaminants. Only Mn had a translocation factor of
over 1, and additionally Cr and Zn in the case of Trifolium (results not shown).
The REE pattern of root normalized to PAAS reflects the general pattern found in soil again,
with a clear Ce positive anomaly and an enrichment of MREE. The difference for the REE
patterns is more marked between shoots and roots than between plants, with for instance a
systematic HREE depletion in shoots compare to roots, due to a different transport in the
plant. There is no comparison possible with previous studies for the fractionation from soil
water to roots, because of the lack of data on normalisation to water. Since the pattern shoed a
wavy structure, it showed that even if the REE coming into the plant were not corresponding
exactly to those present in the soil solution, there was no preferential uptake to HREE or
LREE.
REE can be also be traced within the plant from the roots to the shoots. We could observed
that here also REE behave in a different way. There are too many effects to consider when
studying the fractionation to and within plants, what explains the variety of different effects
described in various articles, and their apparent contradiction. In some studies HREE were
depleted in some cases, in other enriched. Enrichment of MREE has also been reported. Many
effects depend on the plant part considered i.e. if all areal parts, or if the stems are separated
from grains and leaves. Nevertheless, according to previous studies (Lonschinski, M., 2009),
the fractionation of REE within the plant are comparable for same plants on a similar soil, the
pattern still being dependent on the plant species. Few studies deal with the REE fractionation
within plants, so there a few comparisons possible. Nevertheless, despite differences, the most
common reported trend was LREE enrichment in the shoots, as well as the tetrad effect,
which is not visible in our study. Further, Eu anomaly or HREE enrichment in leaves are
101
reported to occur, but were not observed here (Aouad et al., 2006; Liang et al., 2008; Semhi et
al., 2009; Stille et al., 2006).
4.5 Effect of plant consortium Mixed crop cultivation had a positive influence on the appearance of some plants: clover had
bigger leaves; sunflower had shorter shoot growth but slightly healthier colour. So polyculture
seems to have a positive impact on each plant. Furthermore, the concentrations of metals were
lower at mixed crop cultivation than at single crop cultivation. This is especially the case for
Festuca, Trifolium and Triticale. It was also noticeable that the total REE amounts per g dry
biomass were lower at mixed crop cultivation (Figure 2, Tableb), except for Festuca, which is
taking up more metals as polyculture.
Plants seem to take up less metal if grown in a community. Plant can profit from protection
mechanisms of another plant, or of their better nutrient uptake system. For instance, sunflower
can profit from the denser root net given by red fescue that would hold water and retain
metals, and so diminish toxic effects of metals and also its access to nutrients and
consequently reduce growth. Similarly, clover can grow better and produce bigger leaves if
the combination of plants can protect it against metal stress. So, even if the biomass
production of Festuca and Helianthus is better if grown as monoculture, it seems that their
health is still affect by the neighbour plants. It has been reported in effect that plants can
influence each other’s nutrient uptake, as for example peanut facilitates P nutrition of maize
and barley, while maize and barley improve K, Fe, Zn and Mn nutrition (Inal and Gunes,
2008).
102
5 CONCLUSION
Plants influence the rhizosphere zone by changing its pH and the amounts of soluble trace
elements, i.e. their mobility in soil and their uptake. Plants are able to change the soil
properties and many other influencing factors concerning metal mobility. Here we consider
one of the possible mechanisms, the exudation of organic compounds, especially organic
acids. The interaction between plants grown as a consortium is also influencing the amount of
metals taken up, mainly by protecting each other’s from toxic effects of excessive metal
uptake. In particular, REE are mobilised and taken up. The present plants do translocate only
a tenth of the up taken metal into the shoots, with a preference for LREE. The fractionation
observed between soil and soil water with a preferential dissolution of HREE is a hint for the
action of organic substances excreted by plant roots in the rhizosphere zone. Other influences
like the action of microorganisms cannot be excluded. Therefore, the REE signature is a
method to be considered to detect the influence of some organic components in the
rhizosphere, overcoming the analytical limitations.
Figure 10: Overview over the mechanisms influencing REE fractionation and metal uptake in the rhizosphere
Influence of bacteria on lathanide and actinide transfer from specific soil components (humus,
soil minerals and vitrified municipal waste incinerator bottom ash) to corn plants: Sr-Nd isotope evidence. Sci. Total Environ. 370, 545-551.
Aubert, D., Stille, P., Probst, A., 2001. REE fractionation during granite weathering and
removal by waters and suspended loads: Sr and Nd isotopic evidence. Geochim. Cosmochim.
Ac. 65, 387-406.
Bau, M., 1999. Scavenging of dissolved yttrium and rare earth by precipitating iron oxyhydroxide: Experimental evidence for Ce oxidation, Y-Ho fractionation, and lanthanide
tetrad effect. Geochim. Cosmochim. Ac. 63, 67-77.
Blume, H.-P., Brümmer, G., 1991. Prediction of heavy metal behaviour in soil by means of
simple field tests. Ecotox. Eviron. Safe. 22, 164-174.
Bradl, H.B., 2004. Adsorption of heavy metal ions on soils and soils constituents. J. Colloid
P., 2004. Proton release by roots of Medicago murex and Medicago sativa growing in acidic
conditions, and implications for rhizosphere pH changes and nodulation at low pH. Soil Biol.
Biochem. 36, 1357-1365.
Dakora, F.D., Phillips, D.A., 2002. Root exudates as mediators of mineral acquisition in low-
nutrient environments. Plant Soil 245, 35-47.
Díaz-Barrientos, E., Madrid, L., Cardo, I., 1999. Effect of flood with mine wastes on metal
extractability of some soils of the Guardiamar river basin (SW Spain). Sci. Total Environ. 242,
149-165.
Ding, S., Liang, T., Zhang, C., Yan, J., Zhang, Z., 2006. Effects of organic ligants on fractionation of rare earth elements (REEs) in hydroponic plants: an application to the
determination of binding capacities by humic acid for modeling. Chemosphere 65, 1942-1948.
Grayston, S.J., Vaughan, D., Jones, D., 1996. Rhizosphere carbon flow in tees, in comparison
with annual plants: the imporance of root exudation and its impacts on microbial activity and
nutrient availability. Appl. Soil Ecol. 5, 29-56.
Gryschko, R., Kuhnle, R., Terytze, K., Breuer, J., Stahr, K., 2004. Soil extraction of readily
soluble heavy metals and As with 1 M NH4NO3-solution - Evaluation of DIN 19730. J. Soils
and Sediments.
Inal, A., Gunes, A., 2008. Interspecific root interactions and rhizosphere effects on salt ions
and nutrient uptake between mixed grown peanut/maize and peanut/barley in original saline-sodic-boron toxic soil. J. Plant Physiol. 165, 490-503.
Jakubick, A., Jenk, U., Kahnt, R., 2002. Modelling of mine flooding and consequences in the
mine hydrogeological environment: Flooding of the Koenigstein mine, Germany. Environ.
Geol. Water. S. 42, 222-234.
Kahlert, J., 1992. Wismut und die Folgen des Uranbergbaus : eine Tagung der Friedrich-
Ebert-Stiftung am 19. Juni 1992 in Gera, Wirtschaftspolitische Diskurse, Electronic edition ed, Bonn, p. 44.
Karim, M.A., Khan, L.I., 2001. Removal of heavy metals from sandy soil using CEHIXM
gadolinium anomalies in wastewater treatment plant effluents and aquatic environment in the Hérault watershed (South France) Chemosphere 75, 1057-1064.
Römheld, V., Marschner, H., 1986. Evidence for a specific uptake system for iron
phytosiderophores in roots of grasses. Plant Physiol. 80, 175-180.
Semhi, K., Chaudhuri, S., Clauer, N., 2009. Fractionation of rare-earth elements in plants
during experimental growth in vaired clay substrates. Appl. Geochem. 24, 447-453.
Curtobacterium, Sphingomonas and Curtobacterium (Table 3). To our knowledge, there exist
no reported studies about the endophytic population of these plant species on metal
contaminated soil; moreover, in general very little is known about endophytic population in
these species. According to one study (Elo et al., 2000), the diversity in Festuca is high -
about 100 isolates were found. Pseudomonas was the prevalent taxon, with Alcaligenes and
Comamonas, which was not found in humus and might be an obligate endophyte. Further
Arthrobacter, Nocardia; Bacillus and Paenibacillus were specific to this plant according to
the authors. The roots of seem to be also inhabited by spore-forming bacilli and nitrogen
fixers, from the genera Rhodococcus, Paenibacillus and Pseudomonas (Elo et al., 2000).
However, compared with reported rape-, hyperaccumulator plant- or trees-associated bacterial
populations on metal contaminated substrates (Aouad et al., 2006; Porteous Moore et al.,
2006; Sheng et al., 2008; Shin et al., 2012; Ulrich et al., 2008), we can conclude that most of
the genera were isolated before for other plants, whereas the genera Cellulosimicrobium and
Curtobacterium are less common among the endophytes isolated from plants; Olivibacter sp.,
known as a soil bacterium (Wang et al., 2008), was only in our work reported as a plant
endophyte. The differences in the observed endophytic populations suggest that besides
possible variations due to changes in isolation techniques, isolation media and identification
procedures, the environmental conditions in which the plants are growing and most of all the
plant species are determining factors for the composition of the community. Indeed, plants are
described to select specifically which bacterial community would develop in their tissues;
(Wang et al., 2008) showed how the community changed from soil to the roots of different
plants, with in particular a shift from a majority of gram positive strains in the soil to a
dominance of gram negative ones in the plants, and how some genera were found in Festuca
and not in Betula. About 100 isolates were found in Festuca, which is a higher diversity than
in humus (~90). Pseudomonas was the prevalent taxon, with Alcaligenes and Comamonas
(not found in humus). Further strains found belonged to the genera Arthrobacter, Nocardia;
Bacillus and Paenibacillus. The authors suggest that plants species select bacteria associated
with the roots from the bacterial pool in the soil, probably through the production of different
root exudates.
Trifolium roots seem predominantly colonised by proteobacteria (Figure 2), although a large
majority of strains remained unidentified. The stems were also colonised by proteobacteria
and by bacteroidetes in high numbers. Also a large proportion of strains isolated from the
leaves remained unidentified, while the other isolates show higher phylotypic diversity than in
the rest of the plant. Isolates found in leaves were quite equally distributed through the
different phyla, except of bacteroidetes, that were only observed in the stem.
In Festuca, as for Trifolium, a great number of isolates from roots could not be identified; the
identified strains were in majority Proteobacteria with a dominant beta-proteobacteria
population, although firmicutes and actinobacteria were also found.
119
Shoots are largely dominated by one isolate belonging to the ß-proteobacteria. Further,
actinobacteria were found in a high proportion.
Figure 2: Diversity assessment of isolated endophytic microorganisms for different compartments (roots R, Stems St and leaves L) of Trifolium pratense (T) and Festuca rubra (F), calculated based on the isolated CFU /g
4.3 Compartmentalisation An overview of the extent of compartmentalisation of bacteria residing within Trifolium and
Festuca is given Figure 2. A good knowledge about the localisation of a strain within the
plant is important especially in function of phytoremediation purposes since a successful
bacterially enhanced phytoremediation strategy requires a bacterium located in the plant tissue
where the pollutant’s residence time is longest, in particular if, as in case of organic
pollutants, it needs to be degraded. There seems to be a strong tendency for
compartmentalisation in the various plant tissues, illustrated by Figure 3 (a) and (b), although
this seems to be more pronounced in Trifolium than in Festuca.
Some genera are found in more than one compartment, but for example in Trifolium there was
no common genus observed within roots and leaves; in general, endophytes of roots seem to
be clearly different from those of the aerial parts of the plant. We presume that plants are
capable of favouring the dominance of some specific seed endophytes as obligate endophytes
and that the isolated facultative endophytes systemically colonised the inside the plant via the
rhizosphere soil. Similarly the cultivation-dependent analysis showed that shoots of T.
goesingense hosted different microbial populations, although the genera Sphingomonas were
exclusively found in association with the interior of shoots (Rajkumar et al., 2009). This was
the case in the present study for Trifolium; in the case of Festuca, only roots contained
isolates related to Sphingomonas.
The hypothesis, that strains would primarily colonise roots and then further move up towards
the leaves (Lodewyckx et al., 2002), does not seem to be confirmed, since the bacterial
diversity seems to be higher in the aerial parts of the plants, and additionally the composition
of the endophytes is different, another argument in favour of the presence of seed endophytes.
In other studies, (Barzanti et al., 2007) higher numbers of ARDRA types were observed in
roots compared to stems and leaves in the Ni-accumulator plant A. bertolonii. Therefore, it is
probable that leaf endophytic populations are a combination of some bacteria translocated
from the stem, but the majority entering through leaf wounds or stomata, the second
Rhodococcus, Alcaligenes, Ralstonia, some firmicutes (Paenibacillus), cyanobacteria as
Nostoc sp. (Franche et al., 2009; Wang et al., 2008). So it is likely that numerous isolates
show this capacity, even though nitrogen fixation has not been tested in our study. This is of
great importance when using the isolated bacteria for biotechnological purposes. In fact,
additionally to the fact that they are related to biotechnologically interesting strains, the
majority of the isolates found in both autochthonous plants was found to be able to produce
auxin, organic acids and siderophores in vitro. Many were also resistant to different heavy
metals.
As a consequence, these strains are likely to (1) survive in a metal contaminates environment
and (2) improve plant survival and growth under these sub-optimal conditions. However, a
study of (Peterson et al., 2006) notify - with endophytic bacteria from soybean as an example
- about making ecological implications from experiments conducted under typical laboratory
conditions and of the additional roles that well-characterised microbial products may play in
microbial interactions. Furthermore, Sturz and Christie (1996) showed that beneficial strains
123
for one plant turned out to be damaging for another species, showing the “clover-maize”
syndrome; therefore, one should carefully choose the adapted inoculants. So the knowledge
about some bacterial physiological properties is helpful but does give only a limited idea
about what is going on in plantae.
We could select some strains which combined many of these promising properties, as
potential plant growth promoters for bioremediation of sites contaminated with heavy metals.
They are listed in Table 5. They should be tested in future experiments for their actual growth
promoting potential on contaminated soil.
Table 5: Endophytic isolates showing potential for use as plant growth promoter on heavy metal contaminated soil. Production of OA: Organic Acids; Sid: Siderophores; Aux: Auxin; Me: Metal resistance; () little; + much
T., 2012. Characterization of lead resistant endophytic Bacillus sp. MN3-4 and its potential for promoting lead accumulation in metal hyperaccumulator Alnus firma. J. Hazard. Mater. 199–
200, 314-320.
Sturz, A.V., Christie, B.R., 1996. Endophytic bacteria of red clover as agents of allelopathic
Takeuchi, M., Sakane, T., Ynagi, M., Yamasato, K., Hamana, K., Yokota, A., 1995.
Taxonomic Study of Bacteria Isolated from Plants: Proposal of Sphingomonas rosa sp. nov., Sphingomonas pruni sp. nov., Sphingomonas asaccharolytica sp. nov., and Sphingomonas
mali sp. nov. Int. J. Syst. Bacteriol. 45, 334-341.
Tervet, I.W., Hollis, J.P., 1948. Bacteria in the storage organs of healthy plants.
solubilzing and nitrogen-fixing bacteria on solubilization of rock phosphate and their effect on
growth promotion and nutrient uptake by walnut. Eur. J. Soil Biol. 50, 112-117.
Zhang, Z., Schwartz, S., Wagner, L., Miller, W., 2000. A greedy algorithm for aligning DNA
sequences. J. Comput. Biol. 7, 203-214.
127
Chapter 4
CHARACTERISATION OF BACTERIA USED FOR
INOCULATION
The present chapter is composed of the work protocols and the results of additional
characterisation tests.
128
Chapter 4: Characterisation of bacteria used for inoculation
The selected strains were further characterised physiologically. They were therefore grown on
different media in order to test their ability to degrade certain compounds and produce
specific products.
Table 1: List of the 11 selected strains, and their characteristics, that were used to choose them
IAA=Auxin Indole-3 acetic acid; Metal resistance I&II: concentration levels, see previous chapter; 0-2: relative metal
resistance capacity (0 no growth, 1 poor growth, 2 normal growth)
Org Acids Siderophores
IAA
Metal resistance I
Metal resistance II
-Fe +Fe Ni Cd Zn Mn Al Ni Cd Zn Mn
A=6d + + - +++ 1 0 2 2 0 0 0 0 2
F=11d + + - ++++ 0 0 2 2 0 0 0 0 2
B=23c (+) + - +++ 2 2 2 2 0 0 0 0 2
C=26e - + - +++ 2 2 2 2 0 0 0 0 2
I=38d + + - +++ 0 2 2 2 0 0 0 0 2
J=40b - + - +++ 0 0 0 2 0 0 0 0 2
G=50d - + - +++ 2 0 1 2 0 0 0 0 2
H=60d + + - ++ 0 0 2 2 0 0 0 0 2
D=61e + + - +++ 2 0 2 2 0 0 0 0 2
E=62c + + - +++ 2 0 2 2 0 0.5 0 0 2
K=46a + - - +++ 0 0 2 2 0 0 0 0.5 2
1 MATERIALS AND METHODS The collection of these tests is taken from the test series of the Bunte Reihe, after (Schlegel,
1992). Some fast test with young cultures grown 24 h on standard 1 medium, were done.
1.1 KOH fast test for Gram-positive and –negative distinction A simple, rapid method utilising a 3% solution of potassium hydroxide to distinguish between
gram-positive and gram- negative bacteria was applied:
Two drops of a 3% solution of potassium hydroxide were placed on a glass slide. A 2-mm
loop full of bacterial growth, obtained from a 24 h culture on standard I agar, was stirred in a
circular motion in the KOH solution. The loop was occasionally raised 1 to 2 cm from the
surface of the slide. The KOH solution characteristically became very viscous and mucoid
with gram-negative bacteria. A string of the mixture would follow the loop when it was
raised. The KOH test was only considered positive if stringing occurred within the first 30 s
of mixing the bacteria in the KOH solution.
In case gram-positive bacteria are suspended in the KOH solution, no slime should be formed.
Several species of anaerobic bacteria display variable Gram stain reactions which often make
identification difficult. Some strains of Clostridia, Eubacteria, and Bifidobacteria stained
gram negative or gram variable; the KOH test correctly classified these strains as gram-
positive. The KOH test incorrectly grouped some strains of Bacteroides sp., Fusobacterium
sp., Leptotrichia buccalis, and Veillonella parvula, but all Gram stain results for these strains
were consistent for gram-negative bacteria (Halebian et al., 1981).
129
A
B
C
E
D
F Control
I
J
K
G
H
1.2 Catalase Two drops of a 3% solution of H2O2 were placed on a glass slide. A 2-mm loop full of
bacterial growth, obtained from a 24-h culture on standard I agar, was stirred in a circular
motion in the solution. Bubbles formation around the needle after rubbing bacterial material
into 3% H2O2 show O2 formation, and so the activity of the enzyme catalase.
Catalase is a common enzyme found in nearly all living organisms that are exposed to
oxygen. Important catalase-negative genera are Streptococcus, Leuconostoc, Lactobacillus,
Clostridium, and Mycoplasma. Enterococci, Staphylococci and Micrococci are catalase-
positive. Other catalase positive organisms include Listeria, Corynebacterium diphtheriae,
Burkholderia cepacia, Nocardia, the family Enterobacteriaceae (Citrobacter, E.Coli,
(Production of OA: Organic Acids; Sid: Siderophores; Aux: Auxin; Me: Metal resistance; () little; + strong)
W
X
Y
Z
149
Also, consortia of the strains were prepared, keeping into account the relative proportions of
the strains as isolated. One consortium was composed for each plant and plant organ,
combining strengths and weaknesses of their characteristics; for instance, particularly metal
resistant but low IAA producing strains were combined with high IAA producers. They were
referred as consortia W, X, Y and Z. All treatments were performed in 5 independent
replicates.
2.1.1 Survey of plant growth
The plants were grown for 5 weeks in a growth chamber with a 12 h day/night cycle (T: 22°C
day / 18°C night; light conditions: photosynthetic active radiation at plant level 173 µmol/m2.
s-1
). Germination and growth were controlled daily during the first week and every 3 days
later on. The pots were watered when necessary by spraying distilled water on them till
germination, and afterwards by pouring water into the tray. The general health of the plants,
their height and growth density were monitored regularly over the duration of the experiment.
2.1.2 Pot experiment 2
The pot experiment was repeated with homogenised soil from the contaminated study area in
order to study the metal uptake and changes in solubility.
For this, 1 kg (dry weight) soil was used. For each pot, 120 ml deionised water and 12 mL
inoculum were added as described for experiment 1. The OD of the strains at time of
inoculation was about 0.9. On each pot 0.6 g of seeds were sown. No nutrients were added.
As for the first experiment, Trifolium pratense and Festuca rubra were chosen as test plants,
but they were inoculated with only the bacterial strains, that showed to be promising in
experiment 1, i. e. for clover I, J, C and red fescue I, J, C, W, X, Y, Z. Additionally, the soil
was also inoculated with bacteria without plant seeds, to verify the effect of bacteria alone.
Common garden soil was used as a control grown under optimal soil conditions. The plants
were grown for 2 months in the greenhouse and monitored as described for experiment 1. All
treatments were performed in 5 independent replicates, except the inoculation of bacteria
without plants, which were done in triplicate.
2.1.3 Chlorophyll fluorescence measurements
The plants of the second experiment were transported until the measurement place, and the
leaves to measure were removed from the plant and kept humid during dark adaptation of the
photosystem (30 min) before measurement. The fluorescence was measured with fluorcam
(Photon System instruments, Brno, Czech Republic) and the data analysed with the
corresponding software. 6 Trifolium leaves and 6 Festuca shoots were taken of each
treatments from 2 or 3 different pots, at 3 times points and compared to plants grown in
uncontaminated soil.
2.2 Statistical testing Trace element concentrations (Zn, Cu, Fe, Ca, K, Na, Mg, Cd, Pb) in soils and plant parts
(root, stem and leaf) on one hand, and chlorophyll fluorescence on the other hand were
statistically compared for the plants treated with different bacterial inocula. The significance
of the differences between treatments was tested and confirmed by a one-way ANOVA test
and LSD post hoc analysis, with a confidence of 95%.
150
2.3 BOX PCR genomic DNA profiling/fingerprinting For the BOX PCR, genomic DNA was extracted using Qiagen DNA extraction kit and DNA
was amplified in PCR using the genomic DNA as template and one primer 5’-
CTACGGCAAGGCGACGCTGACG -3’ (Murry et al., 1995). The PCR mixture (50 µl)
contained 1 µl template, 5 µl of 10x High fidelity PCR buffer, 2 µl of 50 mM MgCl2, 1 µl of
dNTP at 10 mM, 0.2 µl of Platinum Taq High fidelity DNA polymerase, 2 µL of 10mM
primer.
The PCR was performed in a Mastercycler gradient (Eppendorf) with a hot start performed at
94°C for 5 min, followed by 35 cycles of 94°C for 1 min, 50°C for 1.5 min, and 68°C for 8
min, followed by a final extension performed at 68°C for 8 min.
2.4 Plant and soil material: sample preparation and analysis After 5 weeks of growth, selected plants were harvested. Evaluation parameters used were the
height of the plants (growth promoting effect of bacteria) as well as the estimated growth
density and health (‘colour’) of the plant. Two pots were taken for isolation, and the controls
for comparison.
The plants were taken out of the sand (5 to 10 Festuca seedlings and 1 to 2 clover for each
isolation), and after measuring root length and evaluating general health, separated into leaves
and roots (and also stems for Trifolium). From these organs, endophytic microorganisms were
isolated.
Additionally, the plant biomass was harvested, carefully cleaned with deionised and
suprapure water in order to remove any soil from the surface, dried, weighted and milled to
powder using a mixer mill (type MM400, from Retsch®). Subsequently, metals were
extracted by a microwave assisted pressure digestion (MARS 5, from CEM corporation,
USA) with 65% HNO3 (Merck, p.a., subboiled); the obtained solution was diluted and
centrifuged to be ready for analysis. The soil samples were dried in porcelain plates at room
temperature at the air until constant weight. The water content was calculated from the loss of
weight.
Four g of dry sieved soil were eluted with 40mL of selective extraction solution. Both pure
deionised water and a 1 M ammonium nitrate solution (Merck) were used as selective
extractants. The suspensions were shaken 24h overhead at about 20 rpm (Overhead shaker-
ELU safety lock, Edmund Bühler). For each experiment, blanks (tubes with only elution
solution) were prepared and treated in the same way.
The samples were centrifuged 15 min at 2500 rpm. 15 mL of each sample were filtered
through a 0.45 µm-celluloseacetate filter. From the remaining solution pH (pH 320, WTW)
and electrical conductivity (LF320, WTW) were measured. The samples were acidified with
suprapure HNO3 (63%) and kept at 4°C until analysis.
Soil samples were also analysed for their total metal content. For this, they were milled and
100 mg were putted into TFM vessels. Subsequently, 4 ml 40% HF and 4 ml 70% HClO4
(both suprapur, Merck) were added. After the mixture stood overnight in closed vessels, the
vessels were tightened and heated up to 180°C within 4 h. The temperature was maintained
for 12 h and then the samples were allowed to cool down. In order to evaporate acids, the
system again was heated up to 180°C for a period of 4 h, this time using a special evaporation
hood. This temperature was kept for 12 h. Then, to the remaining solid sample 2ml HNO3
(65%, subboiled), 0.6 ml HCl (30%, Suprapur, Merck) and 7ml of pure water (Pure Lab Plus,
151
USF) were added and the mixture was dissolved by heating at 150°C for 10 h. The cooled
samples were then transferred to calibrated 25 ml PMP flasks (Vitlab). Finally, the solution
was replenished to 25 ml by the addition of pure water for analysis.
The elemental contents in the samples were analysed by ICP-OES (Spectroflame, Spectro) for
the main elements (Al, Ca, Fe, K, Mg, Mn, Na, P, S) and ICP-MS (X Series II, Thermo Fisher
Scientific) for the trace elements (As, Co, Cu, Cd, Mn, Ni, REE (La-Lu), U, Zn).
3 RESULTS
3.1 Plant growth promoting effects: macroscopic observation of roots and
growth density Obvious differences were noticed between the Festuca seedlings grown in contaminated and
uncontaminated artificial soils. On the uncontaminated soil seedlings showed a three-fold
longer root growth and a much higher density of roots (Figure 1a). Further, a clear
improvement of root length was obtained after inoculation of specific strains (I, J, K) (Figure
1b). The effect of the combination of strains was also strong.
Strains I, J and K showed to be the most promising for both species, leading to increased plant
height (not shown) and plant density (Figure 2). The effect was particularly obvious for
Trifolium, since many seedlings did not germinate or died at very early growth stages with
other inocula.
For Trifolium, the consortia were not very efficient in promoting growth, except of Z, still
showing growth below the density and height achieved after inoculation with strains I and J.
Non-contaminated
soil
Contaminated soil
Non-inoculated
Contaminated soil
Inoculated Strain K Strain J
Figure 1: Root growth differences
(a): root and shoot length of Festuca
rubra after 5 weeks of growth in
contaminated soil (left) and
uncontaminated soil (right). Metal
contamination inhibits root growth.
(b): roots and shoots of Festuca after 5
weeks in contaminated soil inoculated
with strains K and J (K:
Curtobacterium sp.; J: Rhizobium
radiobacter). Root growth is enhanced
by inocula, comparable to growth without contamination.
9
8 7
6
5
4
3
2
1
0
-1
-2
-3
-4
-5
-6
-7
-8
-9 [cm]
152
Figure 2: Growth density for Festuca rubra and Trifolium pratense on contaminated and un-contaminated substrate
over time. Comparison between single inocula (A-K) with contaminated (O) and un-contaminated (O+) control ([a]
and [b]); comparison between consortia (W-Z) with contaminated (O) and un-contaminated (O+)([c] and [d].
For Festuca, the consortia, X, Y and Z seemed to have greater effects (more seedlings) than
W (Figure 2d). At that point, also some samples showed chlorotic leaves, probably due to
metal toxicity. Festuca plants inoculated with consortia showed even a better development
than plants grown on uncontaminated substrate.
The effect of inoculation seems better if inoculated as a consortium: the development of the
plants is better than what could be expected by addition of the effect of the two single strains
composing the consortium; the effect on plant density is even more obvious for strains
without particularly high growth promoting effect (Figure 3).
0
1
2
3
4
5
6
0 10 20 30
See
dlin
gs /
Po
t
Time [days]
A
B
C
D
E
F
G
H
I
J
K
O
O+
0
5
10
15
20
25
30
35
40
45
0 10 20 30
See
dlin
gs /
Po
t
Time [days]
A
B
C
E
F
G
H
I
J
K
O
O+
0
1
2
3
4
5
6
0 10 20 30
Seed
lings
/ P
ot
Time [days]
O+
W
X
Y
Z
O
0
5
10
15
20
25
30
35
40
45
0 10 20 30
Seed
lings
/ P
ot
Time [days]
D
W
X
Y
Z
O
O+
[b] [a]
[c] [d]
153
0
10
20
30
40
50
60
0 10 20 30
Seed
lings
/ P
ot
Time [d]
D
E
X
Figure 3: Growth density for Festuca rubra on
contaminated substrate over time. Comparison of
plants inoculates with single strains D and E with
plants inoculated with the consortium X=D+E
Significant increases of root and shoot dry weights were observed when the soil was
inoculated with bacteria, especially with consortia W, X and Y for Festuca (Figure 4a). The
root to shoot ratio shifts from a ratio of about 0.5 on contaminated un-inoculated soil to over
1.5 for inoculated soils, indicating that the root biomass is even more improved than that of
the shoots. Shoot biomass was more than doubled for Trifolium compared to un-amended
plants, and for Festuca it was more than tripled for strain mixtures W, X and Y. For roots the
trend is similar, hence it is even more pronounced for Festuca, root biomass being 10-fold
increased with consortia W, X and Y. However, despite inoculation, on contaminated soil the
biomass never reached that obtained on an uncontaminated soil with optimal nutrient supply;
nevertheless, for Trifolium almost 80% of the control biomass could be obtained after
inoculation of strains I and C.
(a) (b)
Figure 4: Effect of inoculation on plant biomass. Root and shoot biomass and root to shoot ratio (R/S) of Festuca (a)
and Trifolium (b)
3.2 Recovery of inoculated strains in plants
Many bacterial strains could be isolated from the tissues of the inoculated plants. More
cultivable strains were isolated from Festuca than Trifolium (there were about three orders of
magnitude more); furthermore, strain diversity in Festuca was higher compared to Trifolium.
The number of CFU in roots was higher than in shoots for Festuca (x10-100), CFU till ~5.108
CFU/g plant. For Trifolium, it was the other way around: more CFU were isolated from leaves
and stems than from roots (x10). The samples with highest CFU were those inoculated with
the strains C, I, J, K, W, Y (on contaminated substrate) and A, I (on uncontaminated one)
0
0.5
1
1.5
2
2.5
3
3.5
0
2
4
6
8
10
12
14
16
I J C W X Y Z 0 NC
F F F F F F F F F
R/S
dry
bio
mas
s [g
]
root shoot R/S
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0
0.5
1
1.5
2
2.5
3
3.5
4
I J C 0 NC
T T T T T
R/S
dry
bio
mas
s [g
]
root shoot R/S
154
The bacterial diversity and CFU were similar if the soil was contaminated or uncontaminated,
except for I (more diversity in plants grown on uncontaminated soil).
The isolated strains were compared to the inoculated one(s) based on their BOX PCR patterns
(Figure 5).
The inoculations were successful since strains I, J, A, B, C, D, G could be recovered (see
Table 2) from 9 samples out of 13 inoculated.
Figure 5: Example of BOX PCR patterns of some samples (on the left) compared to the inoculated strains (A till K, on
the right); la stands for the 100bp DNA ladder.
Strain C is recovered from the Festuca root (Fr) sample inoculated with consortium W, and I, J and K are all
recovered in the Festuca root or shoot sample (Fr or Fl) inoculated with consortium Z.
Table 2: Recovery of inoculated in the different plant organs
Inoculum F r l
B √ B -/B C √ -/C -
I - - -
J - - -
K √ - -
W √ B(A,C?)/C -
X √ D D
Y √ G G?
Z √ J, I K
A* √ A -
I* √ I I
Inoculum T r st l
I √ - - I J - - - -
r =roots; l= leaves; st= stems; F= Festuca; T= Trifolium
E C F D A G H K la i B J
WFr ZFr ZFl
la la
0.4
0.5
0.6
0.7
0.8
0.9
4 Weeks 7 Weeks 10 Weeks
ΦPSII Trifolium i J C O NC
3.3 Chlorophyll fluorescence: plant stress The quantum yield of the photosystem II (ΦPSII) is higher when the PSII is working more
efficient. ΦPSII can be correlated to the stress experienced by plants (Lichtenthaler and
Miehé, 1997). A lower efficiency corresponds to higher stress, as we assume that stressed
plants would have a lower capacity to use light. In our experiment, ΦPSII was higher in
Trifolium plants grown on non-contaminated soil than in those grown on contaminated soil
(Figure 6), which means that PSII
functioned better. Moreover, bacterial
inoculation of plants grown on contaminated
soil lead to a quantum yield comparable to
that of plant grown on the uncontaminated
soil. This effect was significant and stable
over the duration of the experiment.
Figure 6: Quantum yield of Photosystem II over
time in Trifolium inoculated with different
bacterial endophytes. NC = uncontaminated
3.4 Analysis of soils: water and ammonium nitrate extractable fractions of
metals Generally the amounts of elements extracted with water are about a factor of 10 lower than
with ammonium nitrate (Figure 7); however, the trends observed between the treatments do
not depend on the extracting agent, but on the considered chemical element. Trifolium shows
generally a higher amount of extractable metals than Festuca for the same treatment (i.e.
strains I, J, C), even in the case there is no significant difference in the control. The
ammonium-nitrate extractable fraction of some metals (Ni, Al, REE) is lower with Festuca
than for bare soil, however only if the plants are inoculated with the strain consortia (Figure
7). Some metals as Zn or Pb do not show any differences in the amount of ammonium-nitrate
and water extractable metals (Figure 7). Some metals as the REE and also Al show a
significant influence of the plant and bacterial treatment (Figure 7). The ammonium-nitrate
extractability and the water extractability of certain metals as REE were reduced by particular
strains (I, J, C), and increased if the bacteria were inoculated to plants. Mn behaves in
opposite way to REE, as for example strains C, I, and J inoculated to plants cause a decrease
of soluble available metals compared to un-inoculated plants.
0
1
2
3
4
5
6
7
I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Zn
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 O* I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Zn
Bacteria
Plant
156
Figure 7: Water (left) and NH4NO
3 (right) extrable metal contents of the soil in µg/g
I, J, C: bacterial inocula; W, X, Y, Z combinations of 2-3 bacterial inocula; F: Festuca rubra , T: Trifolium, 0: no inoculum or no plant; O* un-contaminated control
0
1
2
3
4
5
0 O* I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Al
0
10
20
30
40
50
0 O* I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Al
0
0.01
0.02
0.03
0.04
0 O* I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T T T F F F F F F F F F 0 0 0 0 0 0 0 0
∑REE
0
1
2
3
4
5
I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T F F F F F F F F F 0 0 0 0 0 0 0 0
∑REE
0
10
20
30
40
50
0 O* I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Mn
0
50
100
150
200
I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Mn
0
0.5
1
1.5
2
2.5
3
0 O* I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Ni
0
2
4
6
8
10
I J C I J C W X Y Z 0 O* I J C W X Y Z 0
T T T F F F F F F F F F 0 0 0 0 0 0 0 0
Ni
Bacteria
Plant
Bacteria
Plant
Bacteria
Plant
Bacteria
Plant
157
3.5 Analysis of plants: total metal content The uptake of metals by plant aerial parts depends on the plant species, the soil and the
speciation of the metal itself.
In Festuca, REE, Cr and Al were taken up more if the plant was inoculated with the strains I,
J and Z; Mn was behaving opposite, being taken up in lower amounts in case of inoculation
with I, J, C (Figure 8).
The consortia W, X, and Y, resulted in significantly lower uptake of Ni and Cd (results not
shown). Cu and Co were taken up in lower concentrations for all inoculates while Fe and Zn
showed no significant change at this concentration (results not shown).
Similarly, for Trifolium higher concentrations of REE were found in the shoots in case they
were inoculated with the strains J and C (results not shown); for Fe a slightly increased
concentration was found after inoculation with I (results not shown). For U no significant
differences were noticed, except a slight decrease in case the plant was inoculated with strain
I (results not shown). The other relevant metals (Cd, Al, Ni, Zn, Cu, Co, Mn) were not
influenced by inoculation at the present concentration (results not shown).
(a) (b)
Figure 8: Metal content of shoots of Festuca rubra. (a) Mn; (b) ∑REE
4 DISCUSSION For both species, Festuca rubra and Trifolium rubra, increased metal contents in the growing
substrate lead to significantly decreased germination rate, plant survival and plant growth.
4.1 Root growth The root length was clearly different between the treatments (Figure 1). This can be explained
by the fact that several heavy metals are known to affect especially root development (Barceló
and Poschenrieder, 2002). On the other hand, since many of the bacterial strains possess the
capacity to produce plant growth promoting auxins (see Table 1), the inoculated endophytes
should be able to improve germination and cell division of plant tissues, and enhance in
particular root growth. This is one of the factors which may explain the strong differences we
observed between un-inoculated plants grown on contaminated soil and those inoculated with
the selected endophytes. Other properties of endophytes, as ACC deaminase activity may also
be involved. These properties have in fact been exploited in many studies (Sheng et al., 2008)
to increase the biomass of plants grown under stress.
However, the changes of plant growth parameters seemed to be typically dependent on the
plant species as the changes in the root biomass were essentially noticed for Festuca and to a
lesser extend for Trifolium (Figure 4). The root/shoot ratio calculated by Rönkkö et al. (1993),
0
200
400
600
800
1000
I J C W X Y Z 0 O*
F F F F F F F F F
µg/
g
0
0.2
0.4
0.6
0.8
1
1.2
1.4
I J C W X Y Z 0 O*
F F F F F F F F F
µg/
g
158
showed also a great increase for F. rubra through inoculation of root associated N2-fixers or
free N2 fixers as Frankia, whereas other tested plants did not show this effect, despite
noticeable changes in the total biomass.
4.2 Bacterial colonisation of Festuca and Trifolium
Although the endophytic bacteria isolated from Festuca and Trifolium possess various plant
growth promoting features, the use of microbial inoculations for growth promotion requires a
sufficient level of re-colonisation of the introduced microbes.
The presence of the inoculated strains in the inoculated plants shows that colonisation of the
plant occurred; comparison of the BOX PCR patterns of re-isolated strains with those of the
inoculated ones shows the success of the inoculation and suggests a causal relationship
between the inoculations and the growth promoting effect (Figure 5).
Strains A, B, G originate from roots, and were observed to be more recovered from the roots
of the inoculated plants. This is not surprising since many endophytes indeed do not only
show a host specificity but also an organ specificity. Strains D, G, I, J, K were only recovered
if inoculated in combination with others, hence it seems that these strains have increased rates
of inoculation success if inoculated as a consortium. Not recovered strains (E, H) were or (1)
not able to colonise or (2) no real endophytes. More bacteria were found after harvest than at
inoculation (for example J, in Trifolium, Table 2) illustrating the success of colonisation by
the introduced strains. The poorer recovery of inocula in Trifolium is also probable causing
the poor growth of the plant on the contaminated soil (cf. Figure 2 and Table 2).
The fate of the inoculated strains over time is an important question to take into account;
some experiments (Whiting et al., 2001) suggest that the positive effects on plants are not
necessarily specific to the strains of bacteria added, but that also native bacterial population
can have a strong impact. (van der Lelie et al., 2005) suggest that horizontal gene transfer to
the other present bacteria is a feature that can even be of use to help plant promoting bacteria
with specific properties to establish in an already existing bacterial community. This was
confirmed in a field experiment on a TCE-contaminated site by Weyens et al. (2009).
4.3 Protection against stress (Photosynthesis)
Chlorophyll fluorescence as a tool to estimate plants’ susceptibility to environmental changes
has been used by several authors since many years (Atlassi Pak et al., 2009; Lichtenthaler and
Miehé, 1997). The quantum yield is the only parameter which is significantly affected by
metal stress over the entire period of the experiment (Figure 6). Other typical parameters like
the Fv/Fm show also a tendency to support the conclusion that inoculation protects plants from
stress, although this effect is really visible only during the first month to 7 weeks (Figure 6).
This can be also due to the fact that the method uses the blue-green fluorescence, which is
only representative for a part of the photosynthesis process. (Lichtenthaler and Miehé, 1997)
explain that blue-green fluorescence emission can lead to misinterpretation about the
photosystem efficiency, since long term stress events reduce eventually the carotenoid content
of leaves and as a consequence increase the proportion of blue light emission by the plants.
To verify this effect, the fluorescence should be measured at different wavelengths (690 and
735 nm). If this should be confirmed, we can assume that our treatments reduced the blue-
green fluorescence emission and compensated the loss of chlorophyll and carotenoids of the
leaves during long-term stress.
159
4.4 Inoculation with consortia has more positive effects than single strain
inocula This study shows that the simultaneous inoculation of 2 or 3 strains changes the effect on the
plant. Especially with regard to the root biomass and the density of plants grown under metal
stress, the combination of strains causes a synergetic effect (Figure 3). Strains that were not
very efficient growth promoters in case they were inoculated separately had a strongly
positive effect on plant growth when combined with each other. This could be due to the
combination of different complementary properties. Indeed, not all inocula showed for
example high metal resistance, and so growth together with a resistant one could increase
their survival and consequently their growth promoting effect on the host. Similarly, a strain
may produce other hormones that would increase the growth of the bacterial partner, or
provide nutrients with a system not available for the other bacteria. So, by combining two or
more strains, the likelihood of having good survival mechanisms under difficult conditions is
increased.
Indeed, Schmidt et al. (2005) reported that some not metal resistant strains, could grow near
resistant ones, due to the fact that they produce substances that protect them against heavy
metals. Moreover, it has already been shown that bacteria introduced as vectors into the plant
ecosystem can be responsible for natural horizontal gene transfer to the endogenous
endophytic population (Weyens et al., 2009a). Van der Lelie et al. (2005) even suggested that
horizontal gene transfer to the other present bacteria is more probable than an establishment
of a new strain in an already existing stable community.
This aspect of plant growth promotion by consortia has not yet been extensively considered in
the past, even though it was already mentioned by Kozyrovska et al. (1996), who investigated
simultaneous inoculation of 2 endophytes in the context of growth promotion of agricultural
crops on radionuclide contaminated soil. It was suggested that endophytes could help crops to
grow in unfavourable environments, to avoid radionuclide uptake, and be a good alternative
to agrochemicals. In fact, the use of beneficial bacteria to promote plant growth and health has
been suggested already over 20 years ago for agricultural crops (Davison, 1988), and studied
for several plants in phytoremediation later on (Doty, 2008; Guo et al., 2010; Lodewyckx et
al., 2002; Mastretta et al., 2006; Mastretta et al., 2009; Weyens et al., 2010; Weyens et al.,
2009a; Weyens et al., 2009b) the application being mostly limited to one bacterial species per
host.
On the other hand some studies focused on the effect of plant diversity on soil properties, as
to achieve a stable persistent cover it is important to use a mixed vegetation, and combine
grasses, legumes and trees (Kidd et al., 2009; Tessema, 2011; Vangronsveld et al., 1996).
Based on these considerations, it is important to consider possible synergetic effects (or
antagonistic) with the natural community of soil microorganisms when adding bacteria for in
situ phytoremediation. In particular, the action of mycorrhiza is crucial, these fungi being
studied extensively for remediation improvement on heavy metal contaminated site They are
known to protect physically the roots from intrusion and physiologically from stress due to
too high metal concentrations or nutrient depletion (Adriaensen et al., 2003; Adriaensen et al.,
2005; Krznaric et al., 2009; Schützendübel and Polle, 2002).
It is clear that more investigations are needed to better understand the interactions in these
complex systems. Their potential impact for both agriculture and in bioremediation is very
high.
160
4.5 Mobilisation of metals in the soil
The interactions of organic acids released by roots with the soil solid phase appear to be
among the key processes (Puschenreiter et al., 2005). In particular, these authors suggest that
root activities of accumulators as Thlaspi goesingense, such as the exudation of organic acids
triggered the replenishment of soluble Ni from immobile metal fractions of the soil. Different
substances, like organic acids, siderophores, and other complexing agents, are known to
influence the solubility of metals and their uptake by plants.
The increase in the solubility of metals in the soil can be also linked to the properties of the
bacteria, since they are also able to produce siderophores and other metal-chelating
substances. Metallophores are for instance produced by strains of Pseudomonas and
Enterobacter (Whiting et al., 2001). P. aeruginosa can also allow complexation of REE;
further it is able to extract Fe and Mg (Aouad et al., 2006). Sheng et al. (2008) showed the
influence of some bacteria on the solubilisation of Pb in soil and water by P. fluorescens G10
and Microbacterium sp. G16.
Sheng et al. (2008) noticed that some bacteria facilitate the release of the poorly soluble Pb,
thereby enhancing its uptake by plants. Certain metal resistant bacteria have been shown to
possess several properties than can affect both the toxicity and plant availability of metals
through the production of several complexing agents as siderophores or organic acids (Sheng
et al. 2008; Rajkumar et al. 2009).
4.6 Metal uptake
Microbes and microbial consortia alone and in combination with their plant host, can
influence the plant availability of some metals. In the past, even if bacteria were used to
enhance plant resistance to toxic amounts of metals and by consequence also biomass
production, it was often not clear if the improved phytoextraction was attributable to the plant
itself or to a combination of plant and microbes (Rajkumar et al., 2009). Our study shows that
bacteria alone can lead to a decrease in the soluble phase for some metals as REE on one
hand, but on another hand to an increase if those same bacteria were inoculated into plants.
Metal uptake was influenced by the presence of bacteria (Figure 8), but in different ways
depending on the strain and on the metal itself. Indeed, the increased biomass production after
inoculation could in some cases be due to the metal immobilising effect of the endophytic
bacteria, thereby lowering the internal metal availability and by consequence its toxicity for
the host (Shin et al., 2012).
Although we observed a slight increase in pH in the rhizosphere compared to bare soil, the
changes in mobility of metals do not seem to be necessarily correlated with pH changes.
Many bacterial endophytes which are metal resistant are known to enhance metal uptake by
plants and support their growth at the same time (Rajkumar et al., 2009). Whiting et al. (2001)
for instance reported that the increase in the solubility of Zn in the soil was not due to changes
in pH or was not a function of increased root hair growth. Their study indicates that the
bacteria facilitated the release of Zn from the non-labile phase in the soil, thus enhancing Zn
accumulation by T. caerulescens.
The fact that the samples with a higher content of metals in the soluble (i.e. water and
ammonium nitrate extractable fractions) soil fraction were the same with higher metal content
in plant shoots (Figures 7 and 8) suggests that the treatment influences the solubility of metals
in a sustained process. In particular root exudates are known to play an important role for
continuous plant availability of metals out of the soil (Puschenreiter et al., 2005).
161
High amounts of soluble metals in the soil result in high amounts in the plant, and vice versa.
However, this is dependent on the element and on the plant species. The property of certain
plants to take up preferentially specific metals above others has been described already by
numerous authors (Krämer, 2010) and used for remediation purposes in the case of
hyperaccumulators (Sarma, 2011). The choice of the right plant is important, since some
plants are accumulating some contaminants more than others.
Abou-Shanab et al. 2003a, 2006 in (Kidd et al., 2009) demonstrated that the bacterial-induced
enhancing effect on metal extraction effect was dependent upon the metal concentration of
soils, emphasising thereby the need for site-specific evaluation.
5 CONCLUSION This study provides new insights into the opportunities given by the interaction between
plants and their associated microorganisms when growing on a soil containing heavy metals.
It demonstrates the effectiveness of using inoculations of endophytic bacteria to increase
phytoremediation potential, and the enhanced effects of bacterial consortia. We want to
emphasise in this context the importance to consider synergetic effects or possibly
antagonistic effects with natural soil microorganism communities during in situ remediation
processes.
Microbes and microbial consortia alone and in combination with their plant host, can
influence the solubility of some metals and therefore their availability to plants. Some bacteria
can reduce the soluble (i.e. mobile) metal fraction. The mobile fraction of metals is lower with
plants than for bare soil, indicating the stabilising effect of plants.
We suggest using Festuca and Trifolium in addition to metal extracting plants, in order to
improve soil fertility and as protection against wind and water erosion (dense root network).
Festuca is more influenced by bacteria concerning its root development, so should therefore
get particular attention when it comes to choosing plant communities for remediation.
162
REFERENCES
Adriaensen, K., Lelie, D.v.d., Van Laere, A., Vagronsveld, J., Colpaert, J.V., 2003. A zinc-
adapted fungus protects pines from zinc stress. New Phytologist 161, 549-555.
Dimkpa, C.O., 2009. Microbial Siderophores in Rhizosphere Interactions in Heavy Metal-
containing Environments, Biologisch-Pharmazeutische Fakultät. Friedrich-Schiller University, Jena, p. 135.
Doty, S.L., 2008. Enhancing phytoremediation through the use of transgenic and endophytes.
New Phytologist 179, 318-333.
Geletneky, J.W., 2002. Hydrogeologische/Hydrologische Untersuchung einer Prä-
Flutungssituation am Beispiel des Gessentals im ehemaligen ostthüringischen Uranbergbaugebiet, Chemisch-Geowissenschaftliche Fakultät. Friedrich-Schiller-Universität,
Jena.
Gibson, G., Klinck, J., 2005. Canada's Resilient North: The Impact of Mining on Aboriginal
Communities. Pimatisiwin: A Journal of Aboriginal and Indigenous Community Health 3,
”Improving plant growth on heavy metal contaminated soil using selected endophytic
microorganisms” PRE XI conference, Thessaloniki (Greece), July 3rd -6th 2012
172
Curriculum Vitae
Tsilla Boisselet
Nationality French
Date of birth Place of birth
24.08.1983 Périgueux
Gender female
February 2009- present
PhD position, supported by JSMC Institute for Earth Sciences Burgweg 11, 07743 Jena Friedrich Schiller Universität Jena, Germany in cotutelle with Center for environmental Sciences Universiteit Hasselt, Campus Diepenbeek Agoralaan, building D, 3590 Diepenbeek, Belgium
September 2007- June 2008
Internship, Microbiology Department for Food and Feed Safety, Microbiology Laboratory , isolation, detection and cultivation of mycobacteria Veterinary Research Institute (V.R.I) Hudcova 70 62100 Brno (Czech Republic)
October 2006- August 2007 Master thesis for the degree of Master of science, Grade: 1 “Characterisation and Biocontrol of Blue Sapstain Fungi” Vienna University of Technology (TU Wien) Supervisor : Prof. Kurt Messner
October 2001- December 2007 Master of science (Diplomingenieur) Chemical engineering specialization in Biochemistry, Microbiology and Molecular Biology Vienna University of Technology (TU Wien)
August 2006
BEST-course at Lund University (Sweden) about renewable resources for sustainable production
September 2005 internship Laboratory, Analysis of water and sludge quality Hauptkläranlage Wien Simmering : Main Water treatment plant, Vienna (Austria)
August 2004 internship Institute for Chemistry, Cellulose analysis University of natural resources and applied sciences in Vienna (Austria)
June 2001
A-Level specialized in sciences, passed with distinction Baccalauréat Général Scientifique avec Mention Très Bien
September 1994-June 2001 Secondary school : Collège Anne Frank, Périgueux, France / Lycée Jay de Beaufort, Périgueux, France